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.

168,657 papers · 148 categories

Trend · papers per month

55109164218 · Jun 202019922001200920172026
48 results for Oracle Policy

MAMBA learns policies competitive with multiple conflicting oracles.

problem Learning policies from multiple conflicting oracles in reinforcement learning.
method MAMBA uses a gradient estimator in the style of GAE to optimize policies, leveraging demonstrations from multiple weak oracles.
result MAMBA outperforms the state-of-the-art in learning policies competitive with multiple conflicting oracles.

RPI combines imitation and reinforcement learning to improve policies efficiently.

problem High sample complexity in reinforcement learning.
method Active interleaving between imitation and reinforcement learning, using oracle queries for exploration.
result RPI outperforms existing methods across various domains.

GPE algorithm optimizes nonparametric contextual bandits with efficient regret bounds.

problem Optimizing nonparametric contextual bandits with efficient regret bounds.
method Inspired by Policy Elimination, GPE uses oracle-efficient techniques for nonparametric classes with infinite VC-dimension.
result GPE is regret-optimal for policy classes with integrable entropy, and for larger entropy, it provides an ε\varepsilon-greedy algorithm with matching regret bounds.

Oracle-efficient algorithm for offline RL with partial data coverage.

problem Offline reinforcement learning with partial data coverage and constraints.
method PDOCRL, a primal-dual algorithm with decomposed linear-programming formulation.
result Near-optimal, near-feasible policy with \(\widetilde{\mathcal O}(ε^{-2})\) sample guarantee.

Paper addresses online alignment of large language models under uncertain preference feedback.

problem Online alignment of large language models with misspecified preference feedback.
method Formulates an oracle-robust objective as a worst-case optimization problem for log-linear policies, and develops projected stochastic composite updates.
result Shows that the robust objective admits an exact closed-form decomposition and achieves O~(ε2)\widetilde{O}(\varepsilon^{-2}) oracle complexity.

CJE calibrates cheap LLM judges against an oracle, achieving high accuracy at a fraction of the cost.

problem Inexpensive LLM judges can produce biased rankings, leading to unreliable outcomes.
method CJE uses a small oracle to calibrate cheap scores, then evaluates at scale with valid uncertainty.
result CJE achieves 99% pairwise ranking accuracy at 14x lower cost compared to a 16x oracle/judge cost ratio.

Study examines how uncertainty visualization affects analyst trust in automated classification systems.

problem The impact of uncertainty on analyst trust in automated classification systems.
method Empirical study evaluating different active learning query policies and visualizations.
result Query policy significantly influences analyst trust in automated classification systems.

A quantum reinforcement learning algorithm reduces sample complexity.

problem Quantum reinforcement learning under model-free settings with quantum oracle access.
method Quantum Natural Policy Gradient (QNPG) algorithm replacing random sampling with deterministic gradient estimation.
result QNPG achieves a sample complexity of ildeO(ε1.5) ilde{\mathcal{O}}(ε^{-1.5}) for queries to the quantum oracle, significantly improving classical lower bound.

Study improves CTS's approximation regret for combinatorial bandits.

problem Improving CTS's performance on non-exact oracles.
method Develops a new O(log(T)/Δ)\mathcal{O}(\log(T)/Δ) upper bound for CTS under specific conditions.
result First O(log(T)/Δ)\mathcal{O}(\log(T)/Δ) approximation regret upper bound for CTS.

New RL method reduces sample complexity for large policy spaces.

problem Large-scale RL with unknown optimal policies and state/action spaces.
method Introduces eluder dimension for policy space, proving near-optimal sample complexity.
result Near-optimal sample complexity upper bound that depends linearly on eluder dimension.

PILAF optimizes reward models from human feedback for better policy alignment.

problem Creating accurate reward models from human feedback for policy optimization.
method Policy-Interpolated Learning for Aligned Feedback (PILAF) that explicitly aligns preference learning with maximizing underlying oracle reward.
result PILAF is optimal from both optimization and statistical perspectives, demonstrating strong performance in RLHF settings.

We present a new algorithm for the contextual bandit learning problem, where the learner repeatedly takes one of KK actions in response to the observed context, and observes the reward only for that chosen action. Our method assumes access to an oracle for solving fully supervised cost-sensitive classification problem…

2014-02-04abs ↗pdf ↗

Paper proposes a novel policy distillation method for better order execution in noisy markets.

problem Effective order execution in noisy and imperfect market conditions.
method Policy distillation method to guide reinforcement learning towards optimal trading strategies.
result Significant improvements over various baselines in order execution.

A major challenge in contextual bandits is to design general-purpose algorithms that are both practically useful and theoretically well-founded. We present a new technique that has the empirical and computational advantages of realizability-based approaches combined with the flexibility of agnostic methods. Our algorit…

2018-03-03abs ↗pdf ↗

New method improves solving combinatorial optimization problems with smoothed policies.

problem Solving combinatorial optimization problems repeatedly with varying instances.
method Smoothed policies with controlled random perturbations to linear oracle, leading to differentiable surrogate risk.
result Generalization bound decomposes excess risk into bias, estimation, and optimization components.

The paper automates policy learning for nonlinear welfare criteria using machine learning and debiasing techniques.

problem Learning optimal policies from observational data with nonlinear welfare criteria.
method Modeling a nonlinear welfare criterion with a utility function, estimating propensity scores with machine learning, and using sieve approximations and cross-validation for model selection.
result The proposed policy learning method satisfies oracle inequalities, providing theoretical guarantees on performance.

We study the computational tractability of PAC reinforcement learning with rich observations. We present new provably sample-efficient algorithms for environments with deterministic hidden state dynamics and stochastic rich observations. These methods operate in an oracle model of computation -- accessing policy and va…

2018-03-01abs ↗pdf ↗

New RLHF framework handles general preference oracles without reward functions.

problem Handling general preference oracles without assuming a reward function.
method Developed a minimax game between two LLMs for RLHF under a general preference oracle, focusing on KL-regularized preference.
result Proposed algorithms for efficient offline and online RLHF learning.

A new method learns the optimal pricing map for semiparametric dynamic pricing problems.

problem Optimizing pricing strategies in a semiparametric valuation model with unknown utility and noise.
method Developed a modular policy called ORBIT that uses a scalar pilot index, localizes a benchmark price, and learns a local polynomial approximation of the oracle price map.
result Achieves regret bound of \( \widetilde{O}\big(T^{\frac{2β-1}{4β-3}}+\sqrt{dT}\big) \) for the linear utility model and minimax sharp lower bound.

Develops algorithms for CCBs with non-linear costs, improving safety and performance.

problem Safety constraints in sequential decision making with non-linear arm costs.
method Innovative algorithms using Inverse Gap Weighting (IGW) and online regression oracle.
result Sub-linear regret bounds for C-SquareCB and first-order regret for C-FastCB.

We develop a generic data-driven method for estimator selection in off-policy policy evaluation settings. We establish a strong performance guarantee for the method, showing that it is competitive with the oracle estimator, up to a constant factor. Via in-depth case studies in contextual bandits and reinforcement learn…

2020-02-18abs ↗pdf ↗

Probabilistic Bisection Algorithm performs root finding based on knowledge acquired from noisy oracle responses. We consider the generalized PBA setting (G-PBA) where the statistical distribution of the oracle is unknown and location-dependent, so that model inference and Bayesian knowledge updating must be performed s…

2018-06-30abs ↗pdf ↗

Efficiently plans large MDPs with weak function approximations.

problem Planning in large MDPs with limited function approximation capabilities.
method Uses linear value function approximation with weak requirements and a generative oracle.
result Produces almost-optimal actions for any state with polynomial computation time.

Improved algorithm for contextual bandits with reduced regret.

problem Adversarial contextual bandits with i.i.d. contexts.
method Oracle-efficient relaxation with O(T23(Klog(Π))13)O(T^{\frac{2}{3}}(K\log(|Π|))^{\frac{1}{3}}) regret bound.
result First to improve regret bound and match original bound for stochastic case.

The stochastic multi-armed bandit (MAB) problem is a common model for sequential decision problems. In the standard setup, a decision maker has to choose at every instant between several competing arms, each of them provides a scalar random variable, referred to as a "reward." Nearly all research on this topic consider…

2018-06-04abs ↗pdf ↗

While deep reinforcement learning has successfully solved many challenging control tasks, its real-world applicability has been limited by the inability to ensure the safety of learned policies. We propose an approach to verifiable reinforcement learning by training decision tree policies, which can represent complex p…

2018-05-22abs ↗pdf ↗

Sample inefficiency is a long-lasting problem in reinforcement learning (RL). The state-of-the-art estimates the optimal action values while it usually involves an extensive search over the state-action space and unstable optimization. Towards the sample-efficient RL, we propose ranking policy gradient (RPG), a policy …

2019-06-24abs ↗pdf ↗

New algorithm reduces online learning error for unknown feature distributions.

problem Oracle-efficient hybrid online learning with unknown feature and label distributions.
method Computational efficient online predictor using ERM oracle for finite-VC and fat-shattering classes.
result Oracle-efficient sublinear regret bounds for hybrid online learning with unknown feature generation.

New method removes oracle and reduces memory usage for robust MDPs.

problem Applying robust MDPs in practice due to model estimation and oracle requirements.
method Transformed robust MDPs into an alternative form allowing stochastic gradient methods and model-free approach.
result Sample-efficient algorithm with lower storage requirement and no oracle.

We present a reinforcement learning framework, called Programmatically Interpretable Reinforcement Learning (PIRL), that is designed to generate interpretable and verifiable agent policies. Unlike the popular Deep Reinforcement Learning (DRL) paradigm, which represents policies by neural networks, PIRL represents polic…

2018-04-06abs ↗pdf ↗

Thompson Sampling tackles noisy context in stochastic bandits.

problem Designing an action policy for noisy, corrupted contexts in stochastic bandits.
method Introducing a Thompson Sampling algorithm for Gaussian bandits with Gaussian context noise, adopting an information-theoretic analysis.
result Demonstrates the Bayesian regret of the proposed algorithm concerning the oracle's action policy.