Optimization

Motivation

Every data-driven decision problem faces the same trap. A decision-maker estimates a risk functional $R(P,\ell) = \mathbb{E}_P[\ell(\xi)]$ from a nominal distribution $\hat P_N$ built from $N$ training samples, then optimizes a loss function $\ell$ against $\hat P_N$ instead of the unknown true distribution $P$. Because the optimizer adapts to the noise in $\hat P_N$, the in-sample risk of the optimizer systematically understates its true, out-of-sample risk — a phenomenon Smith and Winkler named the optimizer's curse (Smith & Winkler, Management Science, 2006). The remedy explored here is to hedge against a whole neighborhood of plausible distributions around $\hat P_N$, rather than trusting the point estimate. Kuhn, Mohajerin Esfahani, Nguyen and Shafieezadeh-Abadeh's INFORMS TutORials chapter (2019) develops this neighborhood using the Wasserstein distance, and the present mission formalizes its foundational duality theory: the machinery every later result in the chapter (finite-sample guarantees, elliptical tractability, regularization) builds on.

Setting

Fix a norm $\|\cdot\|$ on a finite-dimensional real vector space $E$ (representing $\mathbb{R}^m$). For $p \in [1,\infty)$, the type-$p$ Wasserstein distance between two Borel probability measures $Q, Q'$ on $E$ is

where $\Pi(Q,Q')$ is the set of couplings of $Q$ and $Q'$ — joint probability measures on $E \times E$ whose marginals are $Q$ and $Q'$. The optimal $\pi$ can be read as a transportation plan moving one pile of dirt ($Q$) into another ($Q'$) at minimum cost, which is why $W_p$ is also called the earth mover's distance; the underlying linear program was formalized by Kantorovich (1942) after Monge's 1781 original.

Given $N$ training samples $\hat\xi_1,\dots,\hat\xi_N$, the empirical distribution is $\hat P_N = \frac1N\sum_{i=1}^N \delta_{\hat\xi_i}$. Centered at $\hat P_N$, the Wasserstein ambiguity set of radius $\varepsilon \ge 0$ is

where $\Xi \subseteq E$ is a closed set known to contain the support of the true distribution. The worst-case risk of a loss function $\ell$ is

and minimizing it over a class of admissible loss functions $\mathcal{L}$ is a distributionally robust optimization problem. $\varepsilon$ measures the estimation error one insures against; a larger ambiguity set gives a more conservative (and more expensive) guarantee.

Formalization targets

Goal — Theorem 7, strong duality

This is the Lagrangian dual of the worst-case risk evaluation problem, with $\gamma$ the multiplier of the Wasserstein constraint $W_p(Q,\hat P_N)\le\varepsilon$: it converts a supremum over an infinite-dimensional space of measures into a one-dimensional minimization of the Moreau-Yosida regularization $\ell_\gamma$. Every tractability result later in the chapter (finite convex reformulations, SDP relaxations) specializes this duality by choosing a loss class for which $\ell_\gamma$ is computable.

Supporting dual representations of $W_p$ — Theorems 1 and 2

These identify $W_p$ as a linear program's strong dual (Theorem 1) and, for $p=1$, specialize it to the Kantorovich-Rubinstein form (Theorem 2), which is what lets the worst-case-risk analysis reason about Lipschitz loss functions directly.

Upper and lower bounds — Theorems 5 and 6

These are the tractable, easily-computed bracket that Theorems 7 and 10 later show is tight in important special cases.

Exact case — Theorem 10

Theorem 5's inequality becomes exact under convexity — the cleanest closing corollary of the duality theory, obtained from Theorem 7 by evaluating the Moreau-Yosida regularization of a convex function explicitly.

Significance

Theorem 7 is the hinge on which the entire computational program of Wasserstein distributionally robust optimization turns: every tractable reformulation in the source chapter (piecewise-concave losses via conic duality, quadratic losses via semidefinite programming, the shrinkage-estimator connection) is obtained by substituting a specific loss class into the right-hand side of Theorem 7 and showing the resulting Moreau-Yosida regularization is computable. Kuhn et al. themselves derive it as a corollary of Blanchet & Murthy (2019) and Gao & Kleywegt (2016) for the empirical case, generalized to Polish spaces by Blanchet & Murthy and Gao & Kleywegt independently — the paper cites [12] and [37] for the general statement. Formalizing it is what makes every later, more computational result in the chapter — the ones a solver is more likely to reach for next — rest on a mechanically verified foundation rather than a citation chain.

Status. The mathematical result is well established (multiple independent published proofs cited above); nothing here is open research. What this mission contributes is the first machine-checked formal statement of the duality theorem and its supporting dual representations (Theorems 1, 2, 5, 6, 10) on the Prove2Me platform — none of Wp's dual representation, the Wasserstein ambiguity set, or the worst-case risk functional exist there prior to this mission (see Formalization scope).

Difficulty

The obvious proof strategy — write down the Lagrangian of the semi-infinite program (6), swap the order of the outer supremum over $Q$ and the inner minimization over the multiplier $\gamma$, and invoke ordinary Lagrangian strong duality — fails because (6) is an infinite- dimensional linear program over measures, not a finite convex program: there is no compact feasible set or Slater point in a form that ordinary finite-dimensional duality applies to directly. The actual proof goes through the dual representation of the Wasserstein distance itself (Theorem 1, which is why it is a prerequisite milestone), reformulating the constraint $W_p(Q,\hat P_N)\le\varepsilon$ via its own dual variables and swapping the resulting sup-inf using minimax theorems for semi-infinite programs, not ordinary Lagrangian duality for finite programs.

Formalization scope

$E$ is a generic finite-dimensional real normed space (NormedAddCommGroup, NormedSpace ℝ, Borel-measurable), representing $\mathbb{R}^m$ with the paper's arbitrary fixed norm as a parameter rather than fixing the Euclidean norm. A coupling is formalized directly via MeasureTheory.Measure.map: π.map Prod.fst = Q ∧ π.map Prod.snd = Q'. Constrained infima/suprema (over couplings, over the ambiguity set, over Lipschitz test functions, over perturbation matrices) use Mathlib's guarded-binder idiom ⨅ x (_ : P x), f x, which correctly returns $\top$ (resp. $\bot$) outside the feasible set rather than a finite junk value.

Two deliberate, disclosed conventions keep the extremal-value definitions faithful without extended-real integration machinery, both recorded in MODERATION_NOTES.md:

1. worstCaseRisk and the dual representations (Theorems 1, 2) are valued in EReal, not ℝ, so an unbounded supremum is recorded as $+\infty$ rather than collapsed to Mathlib's real-valued junk value 0 on an unbounded family.
2. The goal theorem (7) and its Moreau-Yosida regularization restrict the loss function to bounded continuous $\ell$ (BoundedContinuousFunction E ℝ), narrower than the paper's general upper-semicontinuous, $\hat P_N$-integrable loss class $\mathcal{L}$ (Assumption 1). This keeps $\ell_\gamma(\xi) = \sup_{z\in\Xi}\ell(z)-\gamma\|z-\xi\|^p$ a finite real number for every nonempty $\Xi$, so the right-hand side's Bochner integral is well-posed; the milestones (Theorems 5, 6, 10) keep the more general real-valued (not necessarily bounded) loss class, since their statements do not require evaluating a pointwise supremum over $\Xi$.
3. Ξ is required closed in Theorems 5, 6 and 7, matching the paper's own standing assumption (p. 6: "we let $\Xi\subseteq\mathbb{R}^m$ be a closed set that is known to contain the support of $P$") for the whole worst-case-risk framework, which is used silently in the paper wherever a theorem takes $\Xi$ as an argument but was not carried into these theorems' own hypothesis lists in an earlier draft.
4. The goal theorem (7) additionally requires $\hat P_N$ itself supported on $\Xi$ ($\hat P_N(\Xi^c)=0$, the same "supported on $\Xi$" convention ambiguitySet uses for $Q\in\mathcal P(\Xi)$), which the paper's framework presupposes for the nominal distribution throughout §2. Combined with $\ell$ bounded, this makes $\ell_\gamma$ bounded on the full-measure set $\Xi$ (above by $\sup\ell$ unconditionally, below by $\ell(\xi)$ itself via $z=\xi$ for $\xi\in\Xi$), which is what makes the right-hand side's integral genuinely well-posed rather than liable to Mathlib's non-integrable junk value $0$.

There is no trivializing formalization risk from a vacuous hypothesis: Ξ.Nonempty and 0 < N are both required exactly where the paper's own indexing and support assumptions require them, and every extremal value uses the extended-real convention above rather than a convention that would make an inequality vacuously true.

No definition in this mission exists on the platform prior to this series (GET /theorems?q=Wasserstein, q=Kantorovich, q=optimal transport, q=coupling return only unrelated discrete/finite-type constructions); all seven definitions and six theorems are drafted fresh. WassersteinDRO.Duality.wassersteinDistance, .ambiguitySet and .worstCaseRisk are the substrate every later mission in this five-part series (Gelbrich tractability, finite-sample guarantees, regularization, shrinkage estimation) either imports directly or redefines locally per the series' reuse rule.

Selected references

• Kuhn, D., Mohajerin Esfahani, P., Nguyen, V. A., & Shafieezadeh-Abadeh, S. (2019). Wasserstein Distributionally Robust Optimization: Theory and Applications in Machine Learning. INFORMS TutORials in Operations Research, 130–166. https://doi.org/10.1287/educ.2019.0198
• Villani, C. (2009). Optimal Transport: Old and New. Springer. (Cited as [108] for Theorems 1 and 2.)
• Smith, J. E., & Winkler, R. L. (2006). The optimizer's curse: Skepticism and postdecision surprise in decision analysis. Management Science, 52(3), 311–322. https://doi.org/10.1287/mnsc.1050.0451
• Gao, R., & Kleywegt, A. J. (2016). Distributionally Robust Stochastic Optimization with Wasserstein Distance. arXiv:1604.02199.
• Blanchet, J., & Murthy, K. (2019). Quantifying Distributional Model Risk via Optimal Transport. Mathematics of Operations Research, 44(2), 565–600. https://doi.org/10.1287/moor.2018.0936

Motivation

A Markov decision process (MDP) chooses a decision in every period of a dynamic system whose state evolves stochastically in response to the decision, so as to maximize expected discounted return. Firms use this model for inventory, pricing, and advertising decisions that respond to a randomly evolving demand state; engineers use it for maintaining or replacing equipment that degrades stochastically. A recurring practical question is qualitative rather than numerical: does the optimal decision increase with the state — should a firm price higher after a period of strong sales, or replace a machine sooner the worse its observed condition — without having to solve the dynamic program numerically for every instance of the model? Topkis's Chapter 3, Section 3.9 (Topkis, Supermodularity and Complementarity, 2011, building on Topkis [1968]) answers this by isolating the lattice-theoretic structure — supermodularity of the return function and of the transition law — under which monotone optimal policies are guaranteed on structural grounds alone. A closely related but logically independent question was studied earlier by Lehmann [1955], who characterized when a family of distributions is stochastically increasing in a parameter; Topkis generalizes Lehmann's characterization to any property of the parameter dependence whose defining set of functions forms a closed convex cone, of which stochastic monotonicity, supermodularity, and convexity are three instances (Theorem 3.9.1, Corollary 3.9.1). Serfozo [1976] independently develops related conditions for partially observed Markov decision processes, and Amir [1996] and Amir, Mirman, and Perkins [1991] give analogous monotonicity results for other classes of dynamic programming models.

Setting

Fix a finite planning horizon of $k$ periods, $i = 1, \dots, k$. In period $i$ the state $t$ ranges over a set $T_i \subseteq \mathbb{R}^m$; given state $t$, the decision $x$ is restricted to a finite, nonempty set $X_{t,i} \subseteq \mathbb{R}^n$ (finiteness guarantees that an optimal decision always exists — no continuity or compactness argument is used). Write $S_i = \{(x,t) : t \in T_i,\, x \in X_{t,i}\}$ for the set of admissible (decision, state) pairs in period $i$. The (bounded) expected net return of choosing decision $x$ in state $t$, period $i$, is $r_i(x,t)$. A discount rate $\beta \in [0,1]$ gives $\gamma = 1/(1+\beta)$, the value in period $i$ of one unit of return in period $i+1$. Given decision $x$, state $t$, and period $i$, the state $w$ of period $i+1$ is drawn from a distribution $F(x,t,i,\cdot)$ on $\mathbb{R}^m$.

The optimal-value function $f_i(t)$ (present value of acting optimally from state $t$, period $i$, onward) and the decision-value function $g_i(x,t)$ (present value of choosing $x$ in state $t$, period $i$, then acting optimally thereafter) are defined by backward induction from period $k$:

A subset $S$ of a Euclidean space is increasing if it is upward closed under the coordinatewise order. A family of distributions $\{F(a,\cdot) : a \in D\}$ indexed by a parameter $a$ is stochastically increasing on $D$ if the probability $\int_S dF(a,w)$ of every increasing set $S$ is a monotone (non-decreasing) function of $a$ on $D$; when $D$ is a sublattice, the family is stochastically supermodular on $D$ if that same probability is a supermodular function of $a$ on $D$. A real-valued function $\varphi$ on a lattice is supermodular on a set $D$ if $\varphi(a_1) + \varphi(a_2) \le \varphi(a_1 \vee a_2) + \varphi(a_1 \wedge a_2)$ for all $a_1, a_2 \in D$ — the mission series' shared notion, defined once in chunk 02-monotonicity and reused here as Supermodularity.Monotonicity.SupermodularOn.

Formalization targets

Goal — Theorem 3.9.2 (monotone optimal policies)

together with the existence of a greatest and a least optimal decision at every state, each increasing in the state, in every period $i$ — under the hypotheses that $S_i$ is a sublattice of $\mathbb{R}^{n+m}$, $X_{t,i}$ is expanding in $t$, $r_i$ is increasing in $t$ (on sections) and jointly supermodular in $(x,t)$, and $F(x,t,i,\cdot)$ is stochastically increasing in $t$ (on sections) and stochastically jointly supermodular in $(x,t)$, for every period $i$.

This is the weakest of the targets that is still worth stating on its own: it packages four related conclusions (parts (a)–(d) of Theorem 3.9.2) that share one hypothesis set, rather than isolating just the headline monotone-decision claim, because the book's own proof derives all four together and a solver attacking part (c) or (d) needs part (a) and (b) established first.

Supporting milestones

• Lemma 3.9.4: under the monotonicity half of the goal's hypotheses alone (no supermodularity), $f_i(t)$ is increasing in $t$ for every period $i$. This is the induction Theorem 3.9.2 reuses and strengthens.
• Corollary 3.9.1(b): on a sublattice $T$, a family of distributions is stochastically supermodular in $t$ iff $\int h(w)\,dF(t,w)$ is supermodular in $t$ for every increasing $h$. This is what lets the goal's hypothesis "$F(x,t,i,\cdot)$ stochastically supermodular in $(x,t)$" be converted into "$\int f_{i+1}(w)\,dF(x,t,i,w)$ supermodular in $(x,t)$", the step that feeds $g_i$'s supermodularity.
• Theorem 3.9.1: the general closed-convex-cone characterization of which Corollary 3.9.1(b) is the supermodularity instance (monotonicity and convexity are the other two instances, not formalized here since the goal's proof only needs the supermodularity case).

Significance

The result gives a purely structural sufficient condition — no smoothness, convexity of the decision set, or specific functional form — for monotone comparative statics in dynamic optimization: whenever the one-period return and the state transition are individually monotone and jointly supermodular in (decision, state), so is the whole multi-period value function, and so are the optimal decisions. Textbook applications include optimal advertising that decreases with prior-period sales, optimal pricing that increases with prior-period sales, and optimal maintenance of a deteriorating system, none of which need to be re-derived from scratch once the structural hypotheses are checked for the specific return and transition functions at hand.

Formalizing it contributes a genuinely new layer to the mission series and, per the paper's own triage (substrate.md), to the platform's application of Mathlib's probability-kernel infrastructure: a finite-horizon MDP with a Lebesgue–Stieltjes-style transition law and its associated stochastic-dominance vocabulary (stochastically increasing / stochastically supermodular families of distributions) did not previously exist on the platform or in this mission series, and both are reusable beyond this mission by any future formalization of dynamic programming under uncertainty. The proof itself is not open — Topkis [2011] (unchanged from the 1968/1998 original) gives a complete, elementary backward-induction argument — so what this mission produces is the formalization of a known, structurally distinctive proof technique, not a new mathematical result.

Difficulty

The naive argument — "supermodularity of $r_i$ plus supermodularity of the transition kernel obviously gives supermodularity of $g_i$" — breaks exactly at the integral: supermodularity of $(x,t) \mapsto F(x,t,i,\cdot)$ is a statement about the whole family of distributions, not about a single number, so "the transition is jointly supermodular" has to be unpacked into "the probability of every increasing set is jointly supermodular in $(x,t)$" before it says anything about $\int f_{i+1}(w)\,dF(x,t,i,w)$ for a specific function $f_{i+1}$. Corollary 3.9.1(b) is exactly the step that licenses this unpacking, and it is not free: it needs Theorem 3.9.1's closed-convex-cone argument (approximating $f_{i+1}$ from below by increasing step functions and invoking monotone convergence), not a pointwise argument on $r_i$ and $F$ separately. A second place the naive argument fails is at the constraint sets: because $X_{t,i}$ only grows with $t$ rather than being fixed, monotonicity of $f_i$ (Lemma 3.9.4) needs its own induction combining that growth with monotonicity of $g_i$ — supermodularity of $g_i$ alone does not hand you monotonicity of the arg max without it.

Formalization scope

States and decisions are represented as Fin m → ℝ and Fin n → ℝ (finite-dimensional Euclidean coordinate spaces with the coordinatewise/product order, matching the book's own restriction to $\mathbb{R}^m$ and $\mathbb{R}^n$ — no abstract lattice is used where the book itself specializes). Distributions are represented as MeasureTheory.Measure on the relevant coordinate space, with IsProbabilityMeasure supplied explicitly wherever a stochastic-dominance hypothesis is used (the probability of a set is read off as (μ S).toReal, which is only faithful to $\int_S dF$ when μ is a probability measure — an unconstrained arbitrary measure would let .toReal collapse an infinite value to $0$ and make the hypothesis trivially satisfiable, a formalization this mission rules out). Integrands h are required integrable against every measure in the family wherever an integral is asserted to lie in a set V, since the Bochner integral of a non-integrable function is definitionally 0 in Lean/Mathlib and would otherwise make Theorem 3.9.1 and Corollary 3.9.1(b) trivially true. Decision sets $X_{t,i}$ are finite (Finset, not merely a bounded or compact Set) and required nonempty exactly where the book assumes it: this finiteness, not any compactness or semicontinuity argument, is what guarantees an optimal decision exists, and dropping it would silently substitute Chapter 2's compactness-based existence machinery for the different argument this section actually uses. The optimal-value and decision-value functions $f_i, g_i$ are represented as any functions satisfying the two backward-recursion equations that define them, rather than being constructed by explicit backward recursion in Lean; since the equations determine $f_i, g_i$ uniquely from $r_i$ and $F$, this is a faithful reading of "define $f_i, g_i$ by (3.9.1) and (3.9.2)," not a weakening of the theorem. The goal's part (c) is stated using the mission series' InducedSetOrder (the Veinott/strong set order, chunk 01-lattices) and part (d) as the existence of two selection functions (greatest, least optimal decision), each monotone in the state — matching the book's "there is a greatest (least) optimal decision ... and this greatest (least) optimal decision is increasing in $t$." The infinite-horizon stationary extension that the book gives immediately after Theorem 3.9.2 (relying on an unproved citation to Blackwell [1965]) is out of scope for this mission.

Reusable infrastructure: the StochasticallyIncreasingOn/StochasticallySupermodularOn definitions are parametric in the ambient preorder/lattice and in the measure's target dimension, so a future mission on stochastic convexity (Corollary 3.9.1(c), not formalized here) or on Topkis's §3.10 stochastic inventory model (which explicitly depends on §3.9, per the book's own reading-order note) can reuse them without modification. Contributions extending this mission to the infinite-horizon case, or completing Corollary 3.9.1's monotonicity and convexity halves, are welcome.

Selected references

• D. M. Topkis, Supermodularity and Complementarity, Princeton University Press, 2011 (unchanged from the 1998 original), Chapter 3, Section 3.9. DOI: 10.1515/9781400822539.
• D. M. Topkis, "Ordered Optimal Solutions," PhD dissertation / working paper, Stanford University, 1968 (the original source for this section's results).
• E. L. Lehmann, "Ordered Families of Distributions," Annals of Mathematical Statistics 26(3), 1955, pp. 399–419. https://doi.org/10.1214/aoms/1177728487
• R. Serfozo, "Monotone Optimal Policies for Markov Decision Processes," Mathematical Programming Study 6, 1976, pp. 202–215.
• R. Amir, "Sensitivity Analysis of Multisector Optimal Economic Dynamics," Journal of Mathematical Economics 25(1), 1996, pp. 123–141.
• R. Amir, L. J. Mirman, and W. R. Perkins, "One-Sector Nonclassical Optimal Growth: Optimality Conditions and Comparative Dynamics," International Economic Review 32(3), 1991, pp. 625–644.
• D. Blackwell, "Discounted Dynamic Programming," Annals of Mathematical Statistics 36(1), 1965, pp. 226–235. https://doi.org/10.1214/aoms/1177700285

Motivation

Every method in Chapters 2-4 of this series solves a projection or proximal subproblem at every step — a Euclidean projection, or a Bregman-divergence prox-mapping — which can itself be as hard as the original problem when $X$ is a complicated feasible set (a spectrahedron, a flow polytope, a matroid base polytope). The conditional gradient method (Frank & Wolfe, 1956) sidesteps this entirely: instead of a projection, each step calls a linear optimization (LO) oracle — minimize a linear function over $X$ — which is frequently far cheaper (over a spectrahedron, this reduces to a single eigenvector computation; over many combinatorial polytopes, to a greedy algorithm). This is the origin of the modern "projection-free" family of optimization methods widely used at the scale where projections are the bottleneck.

Setting

Fix a nonempty compact convex set $X$ in a real normed space $E$ and a convex $f:X\to \mathbb R$ with $L$-Lipschitz gradient (Eq. (7.1.4)): $\|f'(x)-f'(y)\|_*\le L\|x-y\|$. The classic conditional gradient (CndG) method, Algorithm 7.1, sets $x_0\in X$, $y_0=x_0$, and for $k=1,2,\dots$: calls the LO oracle $x_k\in\arg\min_{z\in X}\langle f'(y_{k-1}),z\rangle$, then sets $y_k=(1-\alpha_k)y_{k-1}+\alpha_kx_k$ for a stepsize $\alpha_k\in[0,1]$, either the fixed schedule $\alpha_k=2/(k+1)$ (Eq. (7.1.9)) or exact line search (Eq. (7.1.10)).

Section 7.1.1.2 extends this to bilinear saddle-point problems, where $f$ itself is the (generally nonsmooth) function $f(x)=\max_{y\in Y}\{\langle Ax,y\rangle-\hat f(y)\}$ (Eq. (7.1.5)) for a compact convex $Y$ and linear operator $A$. Since $f$ is nonsmooth, the method is applied instead to a family of smooth approximations $f_\eta$ built from a strongly convex $\omega$ on $Y$ (Eq. (7.1.21)-(7.1.23)), with the smoothing parameter $\eta_k$ allowed to vary across iterations rather than being fixed in advance.

Formalization targets

Goal — Theorem 7.1

Supporting milestones, in attack order

• Lemma 7.1: the smoothed objective family $f_\eta$ is monotone nondecreasing in $\eta\ge0$ — the one-line fact ($V(y)-D_Y^2\le0$ pointwise) that licenses a variable, decreasing smoothing schedule $\eta_k$ rather than a schedule fixed in advance from knowledge of the target accuracy.
• Theorem 7.2: the saddle-point counterpart of the goal theorem, running the same CndG algorithm on the smoothed gradients $f_{\eta_k}'$ instead of $f'$ directly, with the explicit rate $f(y_k)-f^*\le\frac{2}{k(k+1)}\sum_{i=1}^k[i\eta_iD_Y^2+\frac{\|A\|^2}{\sigma_v\eta_i} \|x_i-y_{i-1}\|^2]$.

Every constant here is exactly the book's; the goal theorem's bound is left in terms of the actual step distances $\sum\|x_i-y_{i-1}\|^2$, not a diameter-based simplification (see Difficulty).

Significance

This mission formalizes the founding convergence result of the entire projection-free family (Frank-Wolfe methods), which has become central to large-scale machine learning precisely because its per-iteration cost can be orders of magnitude below that of a projection-based method on structured feasible sets. Theorem 7.1's specific form — a rate depending on the realized step distances rather than a fixed diameter — is also the more informative, tighter statement (the book's own remarks show it recovers the classical diameter-based $O(LD_X^2/\varepsilon)$ complexity as a corollary, but also explains why the rate can be much better in practice when the iterates settle near an extreme point).

No result matching conditional gradient / Frank-Wolfe methods exists on the platform as of 2026-09-18 (q=Frank-Wolfe and q=conditional gradient both return zero hits — see Prior art in MODERATION_NOTES.md).

Difficulty

The chief formalization difficulty is representing "with the stepsize policy in (7.1.9) or (7.1.10)" faithfully without either restricting to one policy (weaker than the book's stated theorem) or introducing an awkward disjunction of two separate algorithm definitions. The book's own proof resolves this by a single observation used for both policies at once: $f(y_k)\le f(\tilde y_k)$ for $\tilde y_k$ the point the fixed schedule $\gamma_k=2/(k+1)$ would have produced — trivially by equality under (7.1.9), or because $y_k$ is chosen to minimize $f$ over the entire line segment under (7.1.10), of which $\tilde y_k$ is one point. This mission's hyk_le hypothesis states exactly this shared consequence, which is genuinely what the proof uses and genuinely covers both policies, rather than picking one arbitrarily.

A second difficulty is not collapsing $\sum_{i=1}^k\|x_i-y_{i-1}\|^2$ into a diameter bound $kD_X^2$ inside the milestone itself — the book's own remarks perform that substitution as a separate, weaker corollary (Eq. (7.1.19)) after stating Theorem 7.1 in its sharper form; folding the substitution into the goal statement itself would silently prove a different, weaker theorem.

Formalization scope

conditional_gradient_rate and saddle_point_cndg_rate state the LO oracle's exactness (x k ∈ Argmin_{z∈X}⟨fGrad(y(k-1)),z⟩) as a pointwise hypothesis rather than deriving it from IsCompact X via an existence lemma — matching the pointwise-hypothesis convention this series uses throughout for argmin-defined algorithmic steps (chunk 03-deterministic's mirror-descent updates, chunk 04-stochastic's stochastic mirror-descent update). X compact convex is still included as a hypothesis, matching the book's own standing assumption on the problem class, even though it is not itself needed to derive the stated conclusion from the other hypotheses.

smoothed_objective_monotone and saddle_point_cndg_rate realize $f_\eta$/$f$ via sSup of the image of $Y$ under the pointwise saddle-point objective, matching the book's own max_{y∈Y}{...} definition (Eq. (7.1.5), (7.1.23)) directly rather than introducing a separate Def_ file for a "bilinear saddle-point objective" structure — no other item in this mission reuses that definition verbatim, so per this series' convention (no shared substrate bundled into a structure unless reused), it is inlined at each use.

A trivializing formalization this mission rules out: stating the LO oracle via an $\varepsilon$-approximate minimizer ((fGrad (y(k-1))) (x k) ≤ (fGrad (y(k-1))) z + ε for some ε) rather than an exact one — this is explicitly a different, weaker algorithm the book does not analyze in Theorem 7.1/7.2 (the book studies approximate LO oracles separately, later in the chapter, not selected here).

Left out of scope, for time: Theorem 7.7 (the matching lower complexity bound for LO-oracle methods, Eq. (7.1.60)) — formalizing it faithfully requires first modeling the abstract class of "LCP methods" (any algorithm restricted to LO-oracle calls) as a universally-quantified object, a substantially different and more involved formalization task than the two upper-bound convergence theorems selected here; named per Hard Rule 7 rather than approximated. The d(x)=\sum x_i\log x_i entropy-smoothing remark and the primal/primal-dual averaging CndG variants (§7.1.2, not covered by this mission's page range) are likewise not attempted.

Selected references

• G. Lan, First-Order and Stochastic Optimization Methods for Machine Learning, Springer Series in the Data Sciences, Springer 2020, Chapter 7, §7.1.1. https://doi.org/10.1007/978-3-030-39568-1
• M. Frank, P. Wolfe, "An algorithm for quadratic programming," Naval Research Logistics Quarterly, 3(1-2), 1956, pp. 95-110.
• M. Jaggi, "Revisiting Frank-Wolfe: projection-free sparse convex optimization," ICML, 2013 (the modern machine-learning revival of the method).

Motivation

Two-stage stochastic linear programming with recourse models a decision made before uncertainty resolves (the first-stage variables $x$) followed by a corrective decision made after (the second-stage, or recourse, variables $y$). Solving such a program means minimizing $c^{\mathsf T}x + Q(x)$, where $Q(x)$ is the expected cost of the best recourse action given $x$ -- an object defined only implicitly, as the value of an embedded linear program that must be solved (or bounded) for every realization of the uncertain data. Before any algorithm for this problem can be justified -- the L-shaped method, stochastic decomposition, scenario decomposition, all developed in later chapters of Birge & Louveaux, Introduction to Stochastic Programming (Springer, 2011) -- one needs to know that $Q$ is well-behaved enough to optimize over at all: that the feasible region is closed and convex, that $Q$ itself is a finite, Lipschitz, convex function on it, that an optimal solution is actually attained rather than only approached in the limit, and finally what an optimality condition for the resulting nonsmooth convex program even looks like. This mission formalizes exactly that foundational layer, Chapter 3, Section 3.1 of the book.

Setting

Fix natural numbers $n_1, n_2, m_1, m_2$ and a finite scenario count $K$. A two-stage recourse instance consists of first-stage data $A \in \mathbb{R}^{m_1 \times n_1}$, $b \in \mathbb{R}^{m_1}$, $c \in \mathbb{R}^{n_1}$, a fixed recourse matrix $W \in \mathbb{R}^{m_2 \times n_2}$, and, for each scenario $k = 1,\dots,K$, a cost vector $q_k \in \mathbb{R}^{n_2}$, a right-hand side $h_k \in \mathbb{R}^{m_2}$, a technology matrix $T_k \in \mathbb{R}^{m_2 \times n_1}$, and a probability $p_k \ge 0$ with $\sum_k p_k = 1$ (Eq. (1.1)). The first-stage feasible region is $K_1 = \{x \mid Ax = b,\ x \ge 0\}$.

For a fixed $x$ and scenario $k$, the second-stage value is

(Eq. (1.6)), taken as an extended real: $+\infty$ if no feasible $y$ exists, $-\infty$ if the inner program is unbounded below. The expected recourse value is $Q(x) = \sum_k p_k\, Q(x,\xi_k)$ (Eq. (1.3)), combined so that $+\infty + (-\infty) = +\infty$ -- the book's own convention (p. 109): infeasibility in one scenario is treated as fatal even if another scenario is unboundedly favorable. The second-stage feasibility set is $K_2 = \{x \mid Q(x) < \infty\}$, and the deterministic-equivalent objective is $z(x) = c^{\mathsf T}x + Q(x)$ (Eq. (1.2)). For $x$ with $Q(x)$ finite, the subdifferential $\partial Q(x)$ is the set of $\eta$ satisfying $Q(x) + \eta^{\mathsf T}(y-x) \le Q(y)$ for every $y$ (p. 115).

A simple-recourse instance is the special case $W = [I,-I]$: the recourse cost splits as $q = (q^+,q^-)$, and $Q(x)$ decomposes componentwise via the closed form of Eq. (1.9)-(1.10) using the (left- and right-limit) distribution functions $F_i^-, F_i^+$ of each $h_i$.

Formalization targets

Goal -- Chapter 3, Theorem 9 (p. 116)

given that (1.2) has a finite optimal value. This is the KKT-style necessary and sufficient optimality condition for the two-stage recourse LP, and the weakest of the mission's targets in the sense that everything else supports it: convexity and finiteness of $Q$ (Theorem 6) are what make the left-to-right implication meaningful, closedness/convexity of $K_2$ (Theorem 5) makes the feasible region well-posed, and attainment (Theorem 8) is what makes "$x^*$ is optimal" a statement about a point that exists rather than an infimum that may not be reached.

Supporting milestones

• Theorem 5(a) (p. 111): $K_2$ is closed and convex.
• Theorem 6(a) (p. 112): $Q$ is finite on $K_2$, and Lipschitzian and convex there.
• Theorem 8 (p. 115): under boundedness of $K_1 \cap K_2$ or eventual linearity of $Q$ along recession directions, a finite optimal value is attained.
• Corollary 10 (p. 116): Theorem 9 specialized to simple recourse, with $\partial Q(x^*)$ replaced by its explicit componentwise description.

Significance

Theorem 9 is the hinge on which the rest of the book's algorithmic chapters turn. The L-shaped method (Chapter 5) is a cutting-plane scheme whose cuts are literally elements of $\partial Q(x)$; stochastic decomposition and sampling-based methods use the same subdifferential structure with estimated cuts; the differentiable specialization (Eq. (1.14), $c + \nabla Q(x^*) = A^{\mathsf T}\lambda^* + \mu^*$) underlies nonlinear-programming approaches to the smooth case. None of this is meaningful without first knowing $Q$ is convex, finite where it needs to be, and that a minimizer exists to characterize. Formalizing this mission's four milestones from the actual definition of $Q$ as an embedded linear program's value -- rather than assuming these properties -- is exactly the content the book itself proves (or, for Theorem 6, explicitly cites to Wets [1972] and Kall [1976] rather than proving); this mission asks for genuine Lean proofs of Theorems 5, 8, 9 and Corollary 10 from the LP structure of $Q$, and records Theorem 6 as a stated (not re-derived) input, matching the book's own presentation.

Difficulty

The obvious shortcut is to treat $Q$ as an opaque convex function and apply a textbook convex-KKT theorem off the shelf. This fails to capture what Theorem 9 actually is: a statement about the specific function $Q(x) = \sum_k p_k \min_y\{q_k^{\mathsf T}y \mid Wy = h_k - T_k x,\ y \ge 0\}$, built from finitely many parametric linear programs, each of which can be infeasible ($Q(x,\xi_k) = +\infty$) or unbounded ($Q(x,\xi_k) = -\infty$) depending on $x$. Convexity of $Q$ must come from convexity of the value function of a parametric LP in its right-hand side (the book's Theorem 2 argument: a convex combination of optimal solutions at two right-hand sides is feasible, hence suboptimal, at the combined right-hand side) -- not from an assumed hypothesis. Handling $\pm\infty$ correctly is a second, easy-to-miss source of error: the book fixes an explicit, non-default convention ($+\infty$ dominates $-\infty$) for combining per-scenario values, the opposite of the convention Mathlib's own extended-real arithmetic uses, so any formalization that reaches for EReal's built-in addition to aggregate $Q$ silently states a different theorem. Theorem 8's attainment condition is a genuine existence result, not an automatic consequence of convexity: continuity alone does not give attainment on an unbounded feasible region, and the book's own counterexample (Eq. (1.11), a negative-exponential tail with infimum $0$ attained by no finite $x$) shows the boundedness/recession hypotheses are load-bearing.

Formalization scope

The scenario set is modeled as Fin K, a finite discrete random variable, matching Section 3.1b's development; under this model "$\xi$ has finite second moments" (the standing hypothesis of Theorems 4-11 in the general, possibly-continuous case) holds automatically and so does not appear as a separate hypothesis anywhere in this mission. Q(x,\xi_k) is defined as an EReal via sInf of the second-stage LP's feasible objective values -- sInf of the empty set is ⊤, and of a set unbounded below is ⊥ -- and is genuinely derived from that inner minimization rather than assumed convex; this rules out the chapter's trivializing formalization, which the paper-level triage explicitly warns against: taking Q(x) as an opaque convex-function hypothesis instead of deriving its properties from the inner LP's structure. Aggregating the $K$ per-scenario values into $Q(x)$ uses a bespoke bookAdd operation implementing the book's stated convention $+\infty+(-\infty)=+\infty$, since Mathlib's EReal addition is defined with the opposite convention ($\bot+\top=\top+\bot=\bot$). $\partial Q(x)$ is the ordinary subgradient-inequality set for this extended-real-valued function.

Theorem 8's condition (b) is stated with the book's own quantifier structure: the threshold $\bar\lambda$ and the recession value depend on the point $x$ and direction $v$ exactly as written, with no strengthening. Theorem 6(a)'s Lipschitz bound is stated, not derived -- the book itself cites it to Wets [1972] and Kall [1976] without proof -- so a faithful Lean proof of that milestone is expected to remain out of scope for this mission. Corollary 10 similarly takes the closed form of $\partial Q_i(x)$ from Eq. (1.10) as a hypothesis on an abstract $Q$, matching how the book itself uses (1.10) as an already-established fact rather than re-deriving it from the second-stage LP in the corollary's own proof. Theorem 11's subdifferential-decomposition result ($\partial Q(x) = E_\omega[\partial Q(x,\xi(\omega))] + N(K_2,x)$) is deliberately left out of this mission's scope: it is not needed by Theorem 9's own proof, and its normal-cone term would require relatively-complete-recourse machinery this mission does not otherwise need. No prior-art match was found on the platform: VectorSpaceOpt.fenchel_duality and the Luenberger-derived VectorSpaceOpt.generalized_kuhn_tucker / kkt_complementary_slackness family use a differentiable (Gateaux-derivative) or conjugate-function KKT model over general normed spaces, not this chapter's finite-dimensional, possibly-nondifferentiable subgradient formulation over the specific polyhedral set $K_1$, so none is a faithful match and all items here are original drafts.

Selected references

• J.R. Birge and F. Louveaux, Introduction to Stochastic Programming, 2nd ed., Springer Series in Operations Research and Financial Engineering, Springer, 2011. https://doi.org/10.1007/978-1-4614-0237-4
• R.J-B. Wets, "Programming Under Uncertainty: The Equivalent Convex Program," SIAM Journal on Applied Mathematics 14 (1966), 89-105 (Lipschitz continuity of the recourse function, cited by the book as Wets [1972] for the closely related result used in Theorem 6). https://doi.org/10.1137/0114008
• D.P. Walkup and R.J-B. Wets, "Stochastic Programs with Recourse," SIAM Journal on Applied Mathematics 15 (1967), 1299-1314 (finiteness of the recourse function and coincidence of the possibility and expectation feasibility sets, underlying Proposition 3 and Theorem 4). https://doi.org/10.1137/0115113

Motivation

The discrete-time Riccati equation is the central object of linear-quadratic optimal control — the design equation behind LQR/LQG controllers in every modern control stack. Proposition 4.4.1 of Bertsekas, Dynamic Programming and Optimal Control, Vol. I (3rd ed., 2005, §4.1) packages its asymptotic theory: under controllability and observability the Riccati iteration converges to the unique positive semidefinite solution of the algebraic Riccati equation and the resulting closed loop is stable. Alongside it, Lemma 4.2.1 of §4.2 develops K-convexity, the analytical engine behind Scarf's optimality of $(s,S)$ inventory policies — a foundational result of operations research. Neither the Riccati asymptotics nor K-convexity exists in Mathlib.

Setting

Matrices $A \in \mathbb{R}^{n\times n}$, $B \in \mathbb{R}^{n\times m}$, $Q = C^\top C \succeq 0$, $R \succ 0$. The Riccati operator (BertsekasRiccatiMap)

$(A,B)$ is controllable if $[B, AB, \dots, A^{n-1}B]$ has rank $n$ (BertsekasControllablePair); $(A,C)$ is observable if $(A^\top, C^\top)$ is controllable (BertsekasObservablePair). Separately, $g : \mathbb{R} \to \mathbb{R}$ is $K$-convex (BertsekasKConvex, Def. 4.2.1) if $K + g(z+y) \ge g(y) + \tfrac{z}{b}(g(y) - g(y-b))$ for all $z \ge 0$, $b > 0$, $y$.

Target

with $L = -(B^\top P B + R)^{-1} B^\top P A$ — BertsekasDP.riccati_convergence_stability (goal). Milestones: Lemma 4.2.1(a)–(d) (kconvex_of_convex, kconvex_combination, kconvex_expectation, kconvex_sS_structure), culminating in the $(s,S)$ structure theorem for continuous coercive $K$-convex functions.

Significance

The Riccati result is the mathematical license behind steady-state LQR design: it guarantees the design equation has one meaningful solution, that iterating the finite-horizon recursion finds it, and that the resulting feedback is stabilizing. Formally it would seed a Mathlib-adjacent theory of matrix fixed-point iterations, positive semidefinite order, and spectral-radius stability. The K-convexity milestones are self-contained real analysis, each of independent reuse value for inventory theory; part (d) is the engine of $(s,S)$-policy optimality. All results are classical and proved in the book; the formal work is new.

Difficulty

The Riccati proof interleaves monotonicity of $F$ on the psd cone, boundedness from controllability (a steering argument), positivity from observability, and stability extracted from the fixed-point identity via a Lyapunov argument — several pieces of matrix analysis (psd order, congruence, Schur-type manipulations, spectral radius vs. convergence of powers) that must be built or located in Mathlib. The naive route of diagonalizing $A$ fails: nothing is symmetric about $A + BL$. For Lemma 4.2.1(d), the difficulty is that $g$ is not convex: the minimizer structure must come from the K-convexity inequality applied at carefully chosen points, plus continuity and coercivity.

Formalization scope

Real matrices over Fin n; Matrix.PosSemidef/PosDef; matrix inverse is Mathlib's total inverse (zero on singular input — harmless here since $B^\top P B + R \succ 0$ along the relevant iterates, which the proof must establish); convergence in the entrywise topology; eigenvalues via spectrum ℂ of the complexified matrix, all strictly inside the unit circle. Rank-based controllability exactly as Def. 4.1.1. K-convexity is stated for all real $K$; note $K \ge 0$ is forced whenever it is satisfiable ($z = 0$), and the expectation milestone is stated for finitely supported disturbances (integrability automatic).

Selected references

• D. P. Bertsekas, Dynamic Programming and Optimal Control, Vol. I, 3rd ed., Athena Scientific, 2005. (Prop. 4.4.1, Def. 4.1.1, §4.2, Lemma 4.2.1.) http://www.athenasc.com/dpbook.html
• R. E. Kalman, Contributions to the theory of optimal control, Bol. Soc. Mat. Mexicana 5 (1960), 102–119.
• H. Scarf, The optimality of (S, s) policies in the dynamic inventory problem, in Mathematical Methods in the Social Sciences, Stanford Univ. Press, 1960.

Motivation

The Pontryagin Minimum (Maximum) Principle is the fundamental necessary condition of optimal control, in continuous use since 1956 across aerospace guidance, robotics, and mathematical economics. Chapter 3 of Bertsekas, Dynamic Programming and Optimal Control, Vol. I (3rd ed., 2005) develops it from the dynamic programming side: the HJB sufficiency theorem (Prop. 3.2.1), an envelope lemma (Lemma 3.3.1), the Minimum Principle itself (Prop. 3.3.1), and its discrete-time counterpart (Prop. 3.3.2). Mathlib's optimal-control coverage is currently near zero — no HJB equation, no adjoint equations, no maximum principle — which makes this the mission with the largest gap between textbook maturity and formal coverage in the series.

Setting

Minimize, over admissible pairs, the cost

with $f, g, h$ continuously differentiable (BertsekasCTModel). Admissible controls are piecewise continuous on $[0,T]$ — formalized as: bounded image and continuous off a finite set (BertsekasPiecewiseContinuousOn) — and state trajectories are continuous, satisfying the ODE off a finite set (BertsekasCTAdmissibleFrom, parametrized by an arbitrary start $(t_0, \xi)$). The Hamiltonian is $H(x,u,p) = g(x,u) + \langle p, f(x,u)\rangle$ (BertsekasHamiltonian).

Target

For an optimal admissible pair $(u^*, x^*)$: there exist an adjoint $p$ and a constant $c$ with

away from finitely many times — BertsekasDP.pontryagin_minimum_principle (goal). Milestones: Prop. 3.2.1 (hjb_sufficiency_of_continuous), Lemma 3.3.1 (envelope_gradient_lemma), Prop. 3.3.2 (discrete_minimum_principle).

The HJB milestone carries the hypotheses that $f$ and $g$ are jointly continuous — the consequence of the §3.1 standing assumptions that its proof uses. An earlier version without any regularity hypothesis was disproved: with a discontinuous running cost the cost integrand need not be integrable, and the library's integral of a non-integrable function is $0$.

Significance

The Minimum Principle converts an infinite-dimensional optimization into a two-point boundary value problem — the basis of shooting methods and of every "bang-bang" analysis. None of it exists in Mathlib; even the HJB verification theorem would be new. The discrete-time milestone is self-contained multivariable calculus and gives early value; the envelope lemma is reusable well beyond control theory. The results are classical (Pontryagin et al. 1962; the book's Chapter 3); the formal proof of Prop. 3.3.1 will need an honest variational argument — the book's own HJB-based derivation is explicitly informal.

Difficulty

For the goal: the classical proofs go through needle variations and a separation argument, or through regularity of the value function — neither is in Mathlib. The book's derivation assumes differentiability of the optimal value function, which is not a hypothesis of the statement; a formal proof must either supply a rigorous variational argument or add intermediate lemmas as new platform problems (sketching is encouraged). For the HJB milestone, the work is differentiating $t \mapsto V(t, x(t))$ along a trajectory that satisfies the ODE only off a finite set, then integrating.

Formalization scope

States and controls in EuclideanSpace ℝ (Fin n) / (Fin m); gradients in Mathlib's gradient; the ODE and adjoint via HasDerivAt off a finite exceptional set; costs via intervalIntegral. Piecewise continuity includes boundedness of the image, so the cost integrand of an admissible pair is genuinely integrable — the junk-value escape (non-integrable integrand ⇒ integral 0) is closed. Fixed initial state, fixed terminal time, free terminal state; time-independent dynamics (so the Hamiltonian is constant, per the book's remark that time-varying systems lose constancy). $U$ is an arbitrary set — no compactness or convexity is assumed in the goal.

Selected references

• D. P. Bertsekas, Dynamic Programming and Optimal Control, Vol. I, 3rd ed., Athena Scientific, 2005. (Ch. 3.) http://www.athenasc.com/dpbook.html
• L. S. Pontryagin, V. G. Boltyanskii, R. V. Gamkrelidze, E. F. Mishchenko, The Mathematical Theory of Optimal Processes, Interscience, 1962.
• W. H. Fleming, R. W. Rishel, Deterministic and Stochastic Optimal Control, Springer, 1975. https://doi.org/10.1007/978-1-4612-6380-7

Motivation

Pontryagin's principle as presented by Luenberger is the decisive necessary condition in continuous-time optimal control. It converts an optimization over functions into a pointwise comparison of a Hamiltonian, coupled to the original state equation and a backward costate equation. Luenberger derives the minimum-Hamiltonian convention from vector-space multiplier ideas and a first-order comparison principle. Formalization is especially valuable here because the printed theorem contains a standard but consequential regularity oversight: it asserts a condition at every time even though controls are only piecewise continuous and the cost is an integral. This mission preserves the intended theorem while replacing that false pointwise claim by the mathematically canonical almost-everywhere statement.

Setting

Fix t₀ < t₁, a finite-dimensional Euclidean state space OCState n, a Euclidean control space OCControl m, and a permitted-control set Omega. A state-control pair (x,u) is admissible when x t₀ = xInit, the state is absolutely continuous on the interval, the control is almost everywhere strongly measurable and lies in Omega almost everywhere, the differential equation x' = F(x,u) holds almost everywhere in the interior, and the running cost is interval integrable. An optimal pair globally minimizes the interval integral among all admissible pairs.

The dynamics F and running cost ell are continuous jointly in state and control and continuously differentiable in the state variable. Their state derivatives Fx and ellx vary continuously. A global Lipschitz estimate controls changes of F in both state and control. Because an a.e. measurable control need not be bounded, the optimal control is explicitly assumed to have an a.e. norm bound on the compact interval, matching the boundedness inherited from the source's piecewise-continuous model. The operator-valued paths Fx (x₀ t) (u₀ t) and ellx (x₀ t) (u₀ t) are also assumed interval integrable along the optimum. Together these hypotheses provide the measure-theoretic regularity needed for the adjoint and state perturbations. The Hamiltonian uses Luenberger's minimum convention,

Formalization targets

The root VectorSpaceOpt.pontryagin_minimum_principle asserts the existence of an absolutely continuous costate lambda with terminal value lambda t₁ = 0. Almost everywhere it satisfies the weak inner-product form of

and almost everywhere on the control interval it satisfies

Two milestones capture source dependencies. control_state_lipschitz_estimate is the Grönwall stability estimate used on p. 263 to control the state response by the integral distance between controls; it explicitly assumes interval integrability of both the state-difference norm and the control-difference norm, so Mathlib's totalized integral cannot hide a nonintegrable input. adjoint_lagrangian_comparison formalizes §9.6, Proposition 1: under an implicit state equation, differentiability in the state, Lipschitz dependence of the state solution, and an adjoint identity, the objective difference agrees with a frozen-state Lagrangian difference up to an explicit filter-level little-o remainder.

Significance

This is the flagship analytic mission of the continuation. It connects finite-dimensional differential calculus, Bochner integration, absolute continuity, ODE constraints, adjoints, and localized control variations in one reusable theorem. The definitions form a minimal control framework that can support terminal costs, endpoint constraints, and alternative maximum-principle conventions later. The corrected a.e. conclusion also demonstrates a central benefit of formalization: informal conventions about representatives of controls and isolated time values must be resolved before a theorem can be accepted.

The weak inner-product adjoint equation avoids introducing a coordinate transpose and remains invariant under the Euclidean-space representation. That choice makes the result immediately reusable in later vector-space treatments of transversality and endpoint multipliers.

Difficulty

The difficulty is very high. Mathlib supplies finite-dimensional calculus, interval integration, absolute continuity, measure-theoretic almost-everywhere statements, and Grönwall tools, but not an assembled Pontryagin framework. The mission must coordinate a state-solution stability estimate, state differentiability of the dynamics and cost, existence and regularity of the backward costate, and Hamiltonian comparison against arbitrary admissible values. The control is measurable rather than globally continuous, so every pointwise expression must be placed under an a.e. quantifier where appropriate. The proposition milestone additionally requires a precise little-o interface instead of an unnamed asymptotic remainder.

Formalization scope

The proposal covers §9.6, Proposition 1 and Theorem 1, with the regularity inherited from the surrounding discussion made explicit. Both the optimal state and the costate are absolutely continuous. Admissible controls are a.e. strongly measurable, which is a broader measure-theoretic proxy for the source's piecewise-continuous controls and is compatible with integral objectives; the root additionally requires the optimal control to be essentially norm bounded on Icc t₀ t₁, restoring the compact-interval boundedness used by the source. The maps F, ell, Fx, and ellx are jointly continuous; state derivatives are supplied by HasFDerivAt; a uniform Lipschitz bound is stated; and both derivative coefficients along the optimal path are interval integrable. The separate Grönwall milestone requires its state and control norm differences to be interval integrable. The interval is required to have positive length.

There is a documented source erratum. The sentence on printed p. 263 states Hamiltonian minimality for every t, while the proof on p. 264 chooses a neighborhood on which a purported strict violation persists. That step requires continuity at the selected time. Moreover, changing a piecewise-continuous control at a single isolated time changes neither its a.e. class, the state equation, nor the integral cost. Therefore no condition can be forced at an arbitrary jump value. The Lean root uses an a.e. conclusion on Icc t₀ t₁; an alternative source-faithful repair would assert the inequality at every continuity point of u₀. The mission does not claim existence of an optimal pair, compactness of Omega, endpoint constraints, nonsmooth dynamics, or a sufficiency theorem.

Selected references

• David G. Luenberger, Optimization by Vector Space Methods, Wiley, 1969, Chapter 9, §9.6, Proposition 1 and Theorem 1, pp. 262–264, including the printed all-times wording and its proof context. Scan: https://sites.science.oregonstate.edu/~show/old/142_Luenberger.pdf
• Lean community, Mathlib documentation, continuously updated: https://leanprover-community.github.io/mathlib4_docs/ (interval integration, absolute continuity, Euclidean spaces, Fréchet derivatives, ODE estimates, and a.e. measurability).

Two quadratic functions can be compared losslessly. The S-procedure says that, when the constraint is strictly feasible, the implication

holds if and only if a single nonnegative multiplier certifies it as a matrix inequality, $\lambda \begin{bmatrix} F_1 & g_1 \\ g_1^{T} & h_1\end{bmatrix} \succeq \begin{bmatrix} F_2 & g_2 \\ g_2^{T} & h_2\end{bmatrix}$ for some $\lambda \ge 0$. It is a cornerstone of control theory, trust-region methods and robust optimization, and a rare case in which a nonconvex problem has zero duality gap. The route runs through the theory this mission builds from Boyd & Vandenberghe §5.8–5.9 and Appendix B: strong alternatives for convex inequality systems, cone-program strong duality under a generalized Slater condition, semidefinite programming duality, the LMI theorems of alternatives, and the hidden convexity of the joint range of two quadratic forms.

Log-concave functions are the meeting point of convex analysis and probability: densities of Gaussian, exponential, uniform and Wishart distributions are all log-concave, and countless facts of applied probability flow from one structural theorem — integrating out variables preserves log-concavity. This mission builds the convex-analysis spine of Boyd & Vandenberghe's Convex Optimization (Chapters 2–3) — separation and supporting hyperplanes, dual cones, the first- and second-order differential characterizations of convexity, Fenchel conjugacy — and climbs to Prékopa's theorem via the Prékopa–Leindler inequality, a landmark of Brunn–Minkowski theory absent from Mathlib.

How much flow can be sent from a source $s$ to a sink $t$ through a network with arc capacities $u_{ij}\in(0,\infty]$ — and what certifies that no more is possible? This mission formalizes §7.4-7.5 of Bertsimas & Tsitsiklis. The circulation calculus of §7.4 supplies the two structural tools: the flow decomposition theorem (Lemma 7.1 — every nonzero nonnegative circulation is a positive combination $\mathbf{f}=\sum_i a_i\mathbf{f}^i$ of simple circulations with only forward arcs, with integer $a_i$ when $\mathbf{f}$ is integer) and the optimality criterion for the minimum cost network flow problem (Theorem 7.6 — a feasible flow is optimal if and only if there is no unsaturated cycle with negative cost). Section 7.5 then formulates the maximum flow problem ($\max b_s$ s.t. $\mathbf{A}\mathbf{f}=\mathbf{b}$, $b_t=-b_s$, $b_i=0$ for $i\ne s,t$, $0\le\mathbf{f}\le\mathbf{u}$), defines augmenting paths (Definition 7.2: $f_{ij}<u_{ij}$ on forward arcs, $f_{ij}>0$ on backward arcs) and the Ford–Fulkerson algorithm, and proves integer invariance and finite termination for integer capacities (Theorem 7.8). The goal is Theorem 7.10:

(a) if the Ford–Fulkerson algorithm terminates because no augmenting path can be found, the current flow is optimal;

(b) the value of the maximum flow equals the minimum cut capacity

— the archetypal combinatorial min-max theorem, which the book notes can also be read as LP duality (pp. 311-312).

Why the two-item periodic JRP is hard

The joint replenishment problem (JRP) is the multi-item extension of the economic order quantity (EOQ) model. Each item is reordered periodically and pays its own ordering and holding costs, and a joint ordering cost is shared by items ordered at the same time. Before this work no hardness result was known for the periodic JRP (p. 10), even though no exact polynomial-time algorithm was known either (Schulz and Telha, arXiv:2007.09045; Operations Research, 2024, as Integer Factorization: Why Two-Item Joint Replenishment Is Hard).

The paper explains this through number theory. In the periodic JRP the joint cost is incurred every $\gcd(q_1, q_2)$ periods, so the objective contains a greatest common divisor of the decision variables. The paper's Theorem 2 states that, assuming the Riemann hypothesis, the two-item periodic JRP is NP-hard under randomized reductions. The proof composes two reductions:

• Shor's reduction from Partition to RangeDivisor ("does $M$ have a divisor in $[L, U]$?"), given in the paper's appendix and based on Shor (2011); this is where the Riemann hypothesis enters;
• a reduction from RangeDivisor to the two-item periodic JRP, whose correctness is Lemma 3 (pp. 13–14).

This mission formalizes Lemma 3. Mission 4 of this series formalizes the Partition half; missions 1 and 2 formalize the integer-factorization reductions behind Theorem 1.

Setting

A period is a positive integer $q$. An item with ordering cost $K$ and holding cost $H$ costs

per unit time. The two-item periodic JRP (objective (3), p. 10) minimizes, over $q_1, q_2 \in \mathbb{Z}_+$,

As a decision problem (p. 13), it asks whether some $(q_1, q_2)$ has cost at most a threshold $z$.

An instance of RangeDivisor (p. 12) is a triple of integers $M \ge U \ge L > 0$ satisfying

it is a yes-instance when $M$ has a divisor $q$ with $L \le q \le U$.

Given such a triple, the instance $\mathrm{JRP}_{M,U,L}$ sets

and the threshold is $z_{M,L,U} = L + U + 2M^2(M+1)$, where $2M^2(M+1) = \mathrm{EOQ}_{K_2,H_2}(M)$. Once item 2 is fixed at $q_2 = M$, item 1 solves problem (5):

In Lean these objects are JRPHard.Aperiodic.eoq, JRPHard.Periodic.periodicCost, JRPHard.Periodic.coefK2, JRPHard.Periodic.coefH2 (shared with missions 1 and 2), and IsRangeDivisorInstance, HasDivisorIn, coefK0, rangeCost, threshold, obj5 and fConv in the namespace JRPHard.RangeDiv.

Formalization targets

Goal: Lemma 3

For every RangeDivisor instance $(M, L, U)$,

Milestones (the steps of the proof, pp. 14–15)

1. Item 2 is pinned: every plan with $q_2 \ne M$ costs more than $z_{M,L,U}$.
2. $f(q) = LU/q + q$ is a lower bound on problem (5)'s objective, $f(L) = f(U) = L + U$, and $f$ is minimized at $\sqrt{LU}$.
3. $\gcd(q, M) \ne q \Rightarrow \gcd(q, M) \le q/2$.
4. If $\gcd(q, M) \ne q$, the objective of (5) at $q$ is at least $2\sqrt{2LU} > L + U$.
5. A divisor $q \in [L, U]$ of $M$ has objective at most $L + U$.
6. A period with objective at most $L + U$ divides $M$ and lies in $[L, U]$.

Significance

Lemma 3 is the step that carries hardness from number theory to inventory theory. With Shor's result it shows that, under the Riemann hypothesis, no polynomial-time exact algorithm for the two-item periodic JRP exists unless NP problems can be solved in randomized polynomial time. The paper concludes (p. 15) that no algorithm for the two-item problem may be simultaneously exact and polynomial on all inputs, so some concession, such as heuristics, must be made.

The result is proved in the paper; it has not been machine-checked. The mission produces a checked statement of the reduction's correctness with every coefficient explicit, and it corrects two imprecisions in the printed lemma (the threshold and the unspecified "sufficiently large" coefficients of item 2; see below). The objects (eoq, periodicCost) are the same ones used in mission 2 of this series.

Difficulty

Each step is elementary, but the statement mixes integer and real reasoning. The objective of (5) is not monotone or convex in $q$: its value jumps between $LU/q + q$ at divisors of $M$ and at least $2LU/q + q$ elsewhere. The argument must show that the gap between these two curves exceeds the window around $L + U$, and this is exactly what RangeDivisor's condition $L + U < 2\sqrt{2LU}$ provides. Without that condition a non-divisor can be cheaper than $L + U$ (for $L = 1$, $U = 100$, $M = 7$, the period $q = 14$ has objective $100/7 + 14 < 101$). The pinning of $q_2 = M$ also has to beat the slack $M + 2\sqrt{LU}$ against $L + U$, which is where $M \ge U$ is used.

Formalization scope

This mission formalizes the correctness of the reduction behind Theorem 2 (its JRP half, Lemma 3); polynomial running time is not formalized. Theorem 2 itself, Shor's NP-completeness of RangeDivisor under randomized reductions, and "one can construct in polynomial time" are not stated. What remains of the polynomial-time claim is the explicit coefficients and the bound $K_0, K_2, H_2 \le 2M^4$ (bit length at most $4\log_2 M + 2$).

Conventions:

• periods are natural numbers, and every quantified period is assumed positive (Lean's $K/0 = 0$ and $\gcd(0, M) = M$ would otherwise create spurious cheap plans);
• costs are real numbers; $L$, $U$, $M$ and the coefficients are natural numbers cast to $\mathbb{R}$;
• RangeDivisor's condition is stated with Real.sqrt exactly as printed;
• "the optimal value is at most $z$" is the decision form "some plan has cost at most $z$" (p. 13); the minimum is attained, so the two agree.

Two disclosed departures from the printed lemma:

• The threshold. The page writes "the optimal value of $\mathrm{JRP}_{M,U,L} \le L + U$". The two-item cost also contains item 2's term $2M^2(M+1)$, so the literal comparison never holds and would make the lemma false. The threshold is $z_{M,L,U} = L + U + 2M^2(M+1)$, the $z_{M,L,U}$ announced in §3.1.
• The item-2 coefficients. The page chooses "sufficiently large values of $K_2$ and $H_2$, as in … Lemma 2". The statement fixes Lemma 2's values $K_2 = M^3(M+1)$, $H_2 = M(M+1)$ rather than quantifying existentially, which would let any instance satisfy the lemma.

A formalization that drops the positivity of periods, states the threshold as $L + U$ on the full two-item cost, or leaves $K_2, H_2$ existential does not capture Lemma 3 and is ruled out.

No library beyond Mathlib (Nat.gcd, Real.sqrt) is needed. Proofs of the milestones, and alternative proofs of the goal, are welcome.

Selected references

• A. S. Schulz and C. Telha, Integer factorization and Riemann's hypothesis: Why two-item joint replenishment is hard, arXiv:2007.09045v1, 2020; published as Integer Factorization: Why Two-Item Joint Replenishment Is Hard, Operations Research, 2024. https://arxiv.org/abs/2007.09045
• P. W. Shor, An NP-complete variant of factoring, Theoretical Computer Science Stack Exchange, 2011 (version 2011-03-01, as cited by the paper). https://cstheory.stackexchange.com/q/4785
• S. Arora and B. Barak, Computational Complexity: A Modern Approach, Cambridge University Press, 2009. https://doi.org/10.1017/CBO9780511804090

Phase recovery and its two convex relaxations

Phase recovery asks to reconstruct a complex signal $x_0 \in \mathbb{C}^p$ from the moduli of linear measurements, $b = |Ax_0| \in \mathbb{R}^n$, where $A \in \mathbb{C}^{n\times p}$ is known. The problem arises in X-ray crystallography, diffraction imaging and microscopy, where detectors record intensities but not phases (Shechtman et al., 2015). Since the map $x \mapsto |Ax|$ is not linear, the problem is non-convex, and a line of work since 2011 replaces it by semidefinite programs over Hermitian matrices.

Two such relaxations are compared in this mission. PhaseLift (Candès, Strohmer and Voroninski, 2013) lifts $x$ to $X = xx^*$ and asks for a positive semidefinite $X$ with $\operatorname{Tr}(a_ia_i^*X) = b_i^2$; its trace-free version is called Weak PhaseLift. PhaseCut (Waldspurger, d'Aspremont and Mallat, 2015) separates amplitude and phase, $Ax = \operatorname{diag}(b)u$ with $|u_i| = 1$, and relaxes the resulting complex MaxCut-type problem to a semidefinite program in a Hermitian matrix $U$ with unit diagonal. In the noiseless case the two relaxations are equivalent (Corollary 4.3 of that paper). This mission formalizes the paper's comparison of the two in the presence of noise: Theorem 4.7 and the two lemmas of Appendix A behind it.

Setting

Write $\mathbf{H}_n$ for the Hermitian $n \times n$ complex matrices and $\mathbf{H}_n^+$ for the positive semidefinite ones. Let $P = AA^\dagger$ be the orthogonal projector onto the range $\mathcal{R}(A)$, with $A^\dagger$ the Moore–Penrose pseudoinverse. Define the face

the affine section $\mathcal{H}_b = \{V \in \mathbf{H}_n : V_{ii} = b_i^2,\ i = 1,\dots,n\}$, and the trace-norm distance $d_1(V,\mathcal{F}) = \inf_{W\in\mathcal{F}}\|V - W\|_1$, where $\|X\|_1 = \operatorname{Tr}\sqrt{X^*X}$.

After rescaling, PhaseCut minimizes $\operatorname{Tr}(V(I - AA^\dagger))$ over $V \in \mathbf{H}_n^+$ with $\operatorname{diag}(V) = b^2$. By Lemma 4.1 this objective equals $d_1(V,\mathcal{F})$ for $V \succeq 0$, so PhaseCut is the projection problem

Weak PhaseLift, as solved by first-order methods with an $\ell^s$ distance ($s \ge 1$), is the projection problem in the other direction,

Both aim at a point of $\mathbf{H}_n^+\cap\mathcal{F}\cap\mathcal{H}_b$; PhaseCut starts from $\mathbf{H}_n^+\cap\mathcal{H}_b$, Weak PhaseLift from $\mathbf{H}_n^+\cap\mathcal{F}$.