The paper introduces a method for certifying the quality of solutions produced by heuristic solvers for general‑sum extensive‑form games. By relating the Nash gap of a weighted average strategy to external regret and a signed correction term, the authors derive an online certificate that can be computed in linear time per checkpoint. Experiments on the biased Leduc family show that selecting the iterate with the smallest certificate from a portfolio of training‑free regret‑matching solvers achieves high success rates across multiple test sets.
Heuristic solvers for general-sum extensive-form games produce a stream of iterates, and practice reports the NashGap of the returned profile. We ask what a solver can know about its own iterates from...
The paper introduces Regret-Weighted Payoff Sampling (RWPS), a budgeted estimator that selectively simulates only payoff-matrix cells relevant to a Nash equilibrium and uses a surrogate model for the remaining entries. RWPS provides an instance-dependent error bound weighted by the opponent’s equilibrium mixture and a coverage result guaranteeing that, once the deviation-relevant set is simulated, surrogate error does not affect either player’s regret. Experiments on three 21×21 general-sum games, including an asymmetric Colonel Blotto, show that RWPS achieves four to six times tighter bounds than previous methods and outperforms other sampling strategies on the CyGym and ANSG cyber simulators at low budgets.
arXiv:2606. 31023v1 Announce Type: cross Abstract: Hard-constrained sequential decision systems have no certified way to spend the test-time compute of modern AI: executing the multi-step drafts of a learned policy or a frozen LLM forfeits the feasibility guarantee a trusted solver provides, while invoking the solver at every step forfeits the speed the AI offers.
arXiv:2608. 06362v1 Announce Type: cross Abstract: Deciding which of two agents is stronger means playing games until skill outweighs luck, and every game costs money, model inference, or expert time.
arXiv:2608. 14761v1 Announce Type: cross Abstract: At a finite public-chance cut, counterfactual regret minimization (CFR) must choose how many outcomes to evaluate before each regret update.
The paper introduces conservative contextual dueling bandits, which enforce that each displayed item’s cumulative score remains above a specified fraction of a trusted baseline’s score. It presents algorithms—MaxInP-BC for linear utilities and BTIGW-BC for general utilities—that guarantee safety with high probability while achieving sublinear regret. Experiments show these methods satisfy the safety constraint more reliably than unconstrained counterparts, with only a moderate increase in regret.
Contextual dueling bandits learn from pairwise comparisons, the same feedback used to fine-tune language models from human preferences. Standard algorithms explore aggressively and can show users two...
arXiv:2608.14761v2 Announce Type: replace-cross Abstract: At a finite public-chance cut, partial counterfactual regret minimization (CFR) can evaluate successive outcome batches at learner states cha...
arXiv:2606. 06486v1 Announce Type: new Abstract: In this paper, we study regret minimization in repeated games with \emph{adaptive} opponents who can respond based on histories of play.
arXiv:2610.14996v1 Announce Type: new Abstract: Choosing a few robust algorithms or schedules can require evaluating every candidate under many adverse scenarios, even though an agent only needs to i...
arXiv:2609.39123v1 Announce Type: new Abstract: When optimizing an expensive black-box function sequentially, as in hyperparameter optimization, we may want to stop once the best evaluated value is c...
arXiv:2607. 05904v1 Announce Type: new Abstract: Training a language model against its own reference-free judgments (the premise of self-rewarding, self-play, and LLM-as-a-judge pipelines) assumes a model's verdict on a shown answer tracks correctness.