This work proves that the supremum approximation achievable with polynomially many value queries and worst-case constant recourse is 1-1/e-\varepsilon, and determines the exact curvature-dependent threshold $1-(\sqrt2-1)\vartheta$ for weighted coverage with weighted coverage with $O(\varepsilon^{-1})$ recourse.
Abstract
Consistent submodular maximization studies the tradeoff between solution quality and stability when elements arrive over time. For a monotone submodular objective, which models diminishing returns, an algorithm maintains a set of at most $k$ available elements and changes only $O(1)$ elements after each insertion. D\"utting et al. [2025] established a tight $2/3$ approximation with unrestricted computation and a polynomial-time $0.51$ approximation. They left open at STOC 2025 whether efficient algorithms can match the offline $1-1/e$ guarantee. We resolve this problem by proving that the supremum approximation achievable with polynomially many value queries and worst-case constant recourse is \[ \beta=2-\sqrt2\approx0.5858<1-1/e. \] For every $\varepsilon>0$, our randomized algorithm attains $\beta-\varepsilon$ with $O(\varepsilon^{-2})$ changes per insertion. Any fixed improvement requires exponentially many queries before one critical insertion or linear recourse of $\Omega(k)$ changes at that insertion, even with unlimited queries afterwards. This gap quantifies the cost of consistency: the current oracle hides which elements will be needed after an arrival. We also determine the exact curvature-dependent threshold $1-(\sqrt2-1)\vartheta$, attain $1-1/e-\varepsilon$ for weighted coverage with $O(\varepsilon^{-1})$ recourse, and separate the existence of universal future-price certificates from their efficient computation. Our algorithm has a bounded-bit polynomial-time implementation for polynomial-bit rational oracle answers; the lower bound uses only logarithmic-bit rational answers.
This work studies non-monotone SMK and focuses on deterministic algorithms with linear query complexity and obtains a deterministic submodular maximization approximation with bicriteria approximation with O(n\log(1/\varepsilon)/\varepsilon)/\varepsilon^2) queries, improving over the previous $O(n^2/\varepsilon)$ query...
These results give the first approximation guarantees exceeding $1/3$ for non-monotone $k$-submodular maximization under matroid and knapsack constraints under classical support constraints.
Si-Yuan Chen, Shengminjie Chen, Sui-Xiang Gao et al.· 0 citations
It is shown that the problem of maximizing a monotone submodular function over maximum cardinality sets in the intersection of two matroids on a common ground set is approximation-equivalent to Submodular Orienteering in directed graphs and an improved polynomial-time bicriteria approximation is obtained via a local se...
Chandra Chekuri, Lars Rohwedder, Neta Singer et al.· arXiv.org· 1 citation
The first deterministic polynomial-time $(2+\varepsilon)$-approximation for Directed Feedback Vertex Set in tournaments is given, for all $\varepsilon>0$.
The soundness uses a result of Haeupler, Saha, and Srinivasan building on the proof of Moser and Tardos, to upper-bound the probability that a fixed relatively large subset is an independent set after the Moser-Tardos algorithm terminates.
Karlin, Klein, and Oveis Gharan introduced a randomized better-than-$3/2$ approximation algorithm for metric TSP [KKO21] and subsequently established the corresponding improvement in the integrality gap of the subtour-elimination LP [KKO22], with an explicit constant $\varepsilon>1.00000\cdot10^{-36}$. Gurvits, Klein,...
Zhao-Feng Song· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduNov 17, 2026
By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.
Microsoft Research Blog· microsoft.comNov 16, 2026
AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduNov 13, 2026