Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

2605207801,040 · Jun 202019922001200920172026
48 results for Group Policy Optimization

GOPO optimizes large models in Hilbert space, avoiding Kullback-Leibler's curvature.

problem Optimizing large language models with Kullback-Leibler divergence's curvature issues.
method GOPO uses Hilbert space L2(pi_k) with orthogonality constraints and a work-dissipation functional.
result GOPO achieves competitive generalization with stable gradient dynamics and entropy preservation.

GRPO optimizes LLMs with verifiable rewards, amplifying policy success.

problem Improving LLMs' reasoning under verifiable binary rewards.
method Introduces GRPO, analyzes variants of reward normalization and regularization.
result GRPO amplifies policy success, converging to a fixed point exceeding the reference.

SPO optimizes LLMs by eliminating group-based baselines and variance issues.

problem Frequent degenerate groups and synchronization barriers in group-based policy optimization methods.
method Single-stream Policy Optimization (SPO) replaces per-group baselines with a persistent, KL-adaptive value tracker and global advantage normalization.
result SPO converges more smoothly and attains higher accuracy than GRPO, improving maj@32 by +3.4 pp across five math benchmarks.

The paper tackles fair sharing of exploration costs across groups in online learning.

problem Sharing the cost of exploration fairly across multiple groups in online learning.
method The paper introduces the 'grouped' bandit model and uses axiomatic bargaining theory, specifically the Nash bargaining solution, to formalize fairness.
result The paper derives policies that are optimally fair and regret-optimal, showing that regret-optimal policies can be unfair.

SFPO optimizes LLM reasoning by repositioning before updating, improving stability and efficiency.

problem Noisy gradients from low-quality rollouts cause instability and inefficient exploration in on-policy RL algorithms.
method Decomposes each step into three stages: a short fast trajectory, repositioning, and slow correction, preserving the objective and rollout process unchanged.
result SFPO consistently improves stability, reduces rollouts, and accelerates convergence, outperforming GRPO on math reasoning benchmarks.

SAGE enhances reinforcement learning by injecting hints to prevent model stagnation.

problem Sparse rewards cause large language models to stall under relative policy optimization.
method SAGE injects privileged hints during training to increase within-group outcome diversity.
result SAGE consistently outperforms GRPO on 6 benchmarks with LLMs, achieving significant improvements.

Doubly fair dynamic pricing ensures equal prices for different groups over time.

problem Achieving equal prices for different groups in online dynamic pricing.
method Online learning algorithm that balances procedural and substantive fairness.
result Achieves ildeO(T) ilde{O}(\sqrt{T}) regret, zero procedural unfairness, and ildeO(T) ilde{O}(\sqrt{T}) substantive unfairness.

Reinforcement Patching optimizes dynamic sequence patching for efficient time series forecasting.

problem Efficiently learning data-adaptive representations for long-horizon sequence data, especially continuous sequences.
method Reinforcement Patching (ReinPatch) uses reinforcement learning to optimize dynamic patching policies and sequence backbones.
result ReinPatch achieves compelling performance in time-series forecasting compared to state-of-the-art methods.

Active-GRPO improves molecular optimization by actively deciding when to imitate or self-improve.

problem Training robust and efficient molecular optimization models with large language models.
method Active-GRPO combines imitation and reinforcement learning, upgrading references and policies dynamically.
result Improves molecular optimization performance, achieving statistically significant gains.

The paper interprets policy-gradient algorithms using continuation theory.

problem Optimizing nonconvex functions in reinforcement learning.
method Formulates policy optimization as optimization by continuation, interprets policy-gradient algorithms as implicitly optimizing deterministic policies.
result Exploration in policy-gradient algorithms is seen as computing a continuation of the return of the policy.

DFL framework improves action and outcome fairness in policy learning.

problem Fairness in policy learning, especially action and outcome fairness.
method Integrates action and outcome fairness into a multi-objective optimization problem using a lexicographic weighted Tchebyshev method.
result DFL framework improves both action and outcome fairness with minimal value reduction.

Proposes CoPO, a new policy optimization method for competitive games.

problem Designing efficient optimization methods for competitive Markov decision processes.
method Competitive policy optimization (CoPO) approach that exploits game-theoretic nature of competitive games.
result Stable optimization, convergence to sophisticated strategies, and higher scores compared to baseline methods.

Optimistic NPG improves policy optimization in online RL with efficient sample complexity.

problem Limited theoretical understanding of policy optimization, especially in online RL.
method Combines natural policy gradient with optimistic policy evaluation.
result Achieves optimal dimension dependence sample complexity for learning near-optimal policies.

CoDistill-GRPO improves small models in GRPO by distilling knowledge from a larger model.

problem Small models in GRPO struggle with sparse rewards on difficult tasks.
method Simultaneously trains a large and small model using co-distillation and GRPO objectives.
result Significant improvement in small model performance over standard GRPO on mathematical benchmarks.

Study optimizes portfolio allocation policies using off-policy data and constraints.

problem Optimizing portfolio allocation policies under constraints using off-policy data.
method Solves a minimax objective with off-policy estimators and online learning to control constraint violations.
result Constructs near-optimal allocation policies for various regimes of operation and constraints.

Extends reinforcement learning alignment to scalar rewards, improving math reasoning.

problem Designing reinforcement learning algorithms for general LLM alignment.
method Introduces f-GRPO and f-HAL, estimating f-divergences between reward-aligned and unaligned distributions.
result Improves math reasoning RLVR tasks and mitigates reward hacking.

DiCE uses diverse agents to explore and learn, avoiding local minima.

problem Local minima in RL due to limited exploration and correlated behavior.
method DiCE employs a group of heterogeneous agents to explore simultaneously and share experiences, with a diversity regularization mechanism.
result DiCE achieves substantial improvement over baselines in MuJoCo locomotion tasks.

Conventional Learning-to-Rank (LTR) methods optimize the utility of the rankings to the users, but they are oblivious to their impact on the ranked items. However, there has been a growing understanding that the latter is important to consider for a wide range of ranking applications (e.g. online marketplaces, job plac…

2019-02-11abs ↗pdf ↗

Recent policy optimization approaches (Schulman et al., 2015a; 2017) have achieved substantial empirical successes by constructing new proxy optimization objectives. These proxy objectives allow stable and low variance policy learning, but require small policy updates to ensure that the proxy objective remains an accur…

2019-10-09abs ↗pdf ↗

Stabilizes policy optimization with off-policy data using divergence augmentation.

problem Premature convergence and instability in policy optimization with off-policy data.
method Incorporates Bregman divergence between behavior and current policies to ensure safe policy updates.
result Empirically shows better performance in data-scarce scenarios compared to other algorithms.

Study uses causal machine learning to assess coupon campaign impact on retailer sales.

problem Assessing the causal effect of a coupon campaign on retailer sales.
method Causal machine learning algorithms, subgroup analysis, optimal policy learning.
result Only two coupon categories (drugstore and other food) have a significant positive impact on sales.

Develops an anytime-valid framework for optimal policy identification from logged contextual bandit data.

problem Selecting the optimal policy from a candidate policy class while monitoring evidence continuously.
method Constructs a time-indexed set that retains the true optimal policy set uniformly over time.
result The procedure allows the analyst to monitor policy values, eliminate clearly suboptimal policies, and stop at data-dependent times without invalidating inference.

Entropy regularization improves policy optimization in reinforcement learning.

problem Improving policy optimization in reinforcement learning.
method Entropy regularization is introduced to soften the greedy policy towards a more diverse softmax policy, leading to a continuously parameterized algorithm that interpolates between policy gradient and Q-learning.
result An intermediate algorithm can improve performance in reinforcement learning.

PS framework selects best policy from library for CSO problems.

problem Policy selection in CSO with heterogeneous performance across covariate space.
method PS framework constructs library of candidate policies and learns a meta-policy to select the best one.
result PS consistently outperforms best single policy in heterogeneous CSO problems.

Paper addresses offline policy evaluation in RL, achieving near-optimal bounds for various policy classes.

problem Evaluate all policies in a class simultaneously for offline RL.
method Uniform convergence in OPE for various policy classes, achieving optimal episode complexity.
result Achieves optimal episode complexity of O(H^3/d_mε^2) for identifying ε-optimal policies.

Policy optimization is a core component of reinforcement learning (RL), and most existing RL methods directly optimize parameters of a policy based on maximizing the expected total reward, or its surrogate. Though often achieving encouraging empirical success, its underlying mathematical principle on {\em policy-distri…

2018-08-09abs ↗pdf ↗

Presents SPEED, an algorithm for optimal policy evaluation in linear bandits with heteroscedastic noise.

problem Optimal data collection for policy evaluation in linear bandits with heteroscedastic reward noise.
method Formulated an optimal design for weighted least squares estimates, derived the optimal sample allocation, introduced SPEED algorithm, and derived regret bounds.
result SPEED leads to policy evaluation with MSE comparable to oracle strategy and significantly lower than random policy execution.

A new method removes policy optimization in adversarial imitation learning.

problem Adversarial imitation learning's delicate alternated optimization.
method Explicitly condition discriminator on two policies, solving generator's optimization problem directly.
result Simpler approach competitive to prevalent methods.

Policy gradient converges to globally optimal policy in nearly linear-quadratic systems.

problem Finding optimal policies in nonlinear control systems with partial information.
method Policy gradient algorithm designed for nearly linear-quadratic regulators with small Lipschitz nonlinear components.
result Policy gradient algorithm converges to globally optimal policy with linear rate.

New algorithm finds near-optimal policies efficiently in zero-sum games.

problem Lack of provable efficiency guarantees for policy optimization in zero-sum games.
method Policy optimization algorithm with function approximation.
result Proves efficient convergence to near-optimal policies with polynomial samples and iterations.