Study optimal reward schemes for inducing desired player performance in risky contests.
problem Designing optimal rewards to encourage desired performance levels in risky contests.
method Analyzed the optimal reward schemes for inducing average and specific rank performance.
result Optimal reward schemes can have surprising shapes, not just related to inequality.
Reward collapse occurs when ranking-based reward models yield uniform rewards for different prompts.
problem Reward collapse in aligning large language models with human preferences.
method Introduced a prompt-aware optimization scheme to derive closed-form expressions for reward distributions.
result Our prompt-aware utility functions significantly alleviate reward collapse during training.
Paper improves sample complexity for reward-free RL in low-rank MDPs.
problem Reward-free RL in low-rank MDPs with unknown representation and weights.
method Proposes a novel model-based algorithm RAFFLE with improved sample complexity.
result RAFFLE achieves ε ε ε -optimal policy and accurate system identification with significantly fewer samples. Ranked Reward algorithm improves bin packing performance.
problem Improving reinforcement learning for combinatorial optimization.
method Ranking rewards from self-play to create a relative performance metric.
result Ranked Reward algorithm outperforms other methods on bin packing problems.
Framework for optimizing search engine rankings using observational data.
problem Optimizing ranking policies for search engines using limited observational data.
method Formulated expected reward optimization problem, estimated context value distribution, trained ranking policy via Bayesian inference.
result Demonstrated trade-offs in ranking policies trained on empirical reward estimates.
Scalable model for slate recommendation learns reward probabilities.
problem Scalable personalized slate recommendation in large action spaces.
method Probabilistic Rank and Reward (PRR) model combining reward, interaction, and rank.
result PRR outperforms existing methods and is scalable to large action spaces.
New algorithm for fair ranking in contextual bandits with concave rewards.
problem Fair ranking in recommendation systems.
method Geometric interpretation of CBCR as optimization, Frank-Wolfe analyses.
result First algorithm with provably vanishing regret for CBCR.
D-REX learns reward functions from ranked demonstrations to beat the demonstrator's performance.
problem Limited performance of imitation learning compared to the demonstrator.
method Disturbance-based Reward Extrapolation (D-REX) that generates ranked demonstrations from behavioral cloning.
result D-REX significantly outperforms standard imitation learning approaches and beats the demonstrator's performance.
Study reward-free RL in non-linear settings, improving efficiency and removing assumptions.
problem Improving sample efficiency in reward-free reinforcement learning for non-linear function approximation.
method Proposed RFOLIVE algorithm for minimal structural assumptions, analyzed hardness results for reward-free and reward-aware exploration.
result Statistical efficiency and hardness results under various structural assumptions, no need for reachability or explorability assumptions.
Algorithm for low-rank matrix bandits with heavy-tailed rewards, achieving nearly optimal regret bound.
problem Stochastic low-rank matrix bandit with heavy-tailed rewards.
method LOTUS algorithm using truncation and dynamic exploration.
result Regret bound of order $ ilde O(d^ rac{3}{2}r^ rac{1}{2}T^ rac{1}{1+δ}/ ilde{D}_{rr})$ without knowing T T T . Aims to learn optimal behavior from ranked experts in MDPs.
problem Learning optimal behavior without a specified reward function from ranked experts.
method Uses ordinal regression to find a reward function maximizing the margin between ranks.
result Shows important differences in reward function hidden from existing algorithms.
The study categorizes reward errors in reinforcement learning, finding some can be beneficial.
problem Training language models with imperfect proxy rewards.
method Theoretical analysis of policy gradient optimization and categorization of reward errors.
result Reward errors can be benign or even beneficial, preventing policy from stalling.
We implement momentum strategies using reward-risk measures as ranking criteria based on classical tempered stable distribution. Performances and risk characteristics for the alternative portfolios are obtained in various asset classes and markets. The reward-risk momentum strategies with lower volatility levels outper…
T-REX improves reinforcement learning from suboptimal demonstrations.
problem Inability of existing IRL methods to significantly outperform the demonstrator.
method Trajectory-ranked Reward EXtrapolation (T-REX) algorithm.
result T-REX outperforms state-of-the-art methods on Atari and MuJoCo tasks.
The paper addresses calibration in label ranking, a structured prediction task.
problem Calibration in label ranking is not well understood and often poorly calibrated.
method Formalized calibration for label ranking, developed a hierarchy of notions, and empirically evaluated models.
result Popular label ranking models are often poorly calibrated, with differences between sub-ranking and top-k metrics.
Paper introduces G-LowTESTR for efficient tensor bandits.
problem Efficient decision-making in multi-dimensional data with non-linear reward functions.
method Generalized low-rank tensor contextual bandits model and G-LowTESTR algorithm.
result G-LowTESTR achieves superior regret bound compared to vectorization and matricization methods.
A fair policy for hiring candidates from different groups is proposed in a linear contextual bandit problem.
problem Selecting candidates from different sensitive groups in a fair manner.
method A greedy policy that constructs a ridge regression estimate and computes relative rank using empirical cumulative distribution function.
result The greedy policy achieves fair pseudo-regret of order d T \sqrt{dT} d T after T T T rounds, satisfying demographic parity. Introduces tensor bandits for multi-dimensional online decision making.
problem Optimal decision making in multi-dimensional online scenarios.
method Stochastic low-rank tensor bandits, tensor elimination, tensor epoch-greedy, tensor ensemble sampling.
result Tensor elimination and tensor epoch-greedy algorithms outperform existing methods.
The paper tackles pure exploration in multi-armed bandits with low rank structure using oblivious sampling.
problem Pure exploration in multi-armed bandits with low rank reward sequences.
method The approach involves separating the exploration strategy from feedback, using oblivious sampling, and incorporating kernel information of reward vectors.
result Efficient algorithms with regret bound O ( d ( ln N ) / n ) O(d\sqrt{(\ln N)/n}) O ( d ( ln N ) / n ) for both time-varying and fixed cases, with a lower bound gap of O ( ln N ) O(\sqrt{\ln N}) O ( ln N ) . New spectral methods improve matrix estimation in RL with low-rank structure.
problem Estimating matrices with low-rank structure in reinforcement learning.
method Spectral-based matrix estimation approaches.
result Spectral methods efficiently recover singular subspaces and minimize entry-wise error.
New algorithm catches moving subspaces in bandit problems.
problem Adapt to changing low-dimensional latent subspaces in bandit settings.
method Piecewise-stationary low-rank linear contextual bandits with CUSUM-style boundary detection.
result Achieves intrinsic rank dynamic regret rate of O ( r T ) O(r\sqrt{T}) O ( r T ) . RAFT fine-tunes models using high-quality samples to align them with human preferences.
problem Aligning generative models with human ethics and preferences.
method RAFT selects high-quality samples, discards undesired behavior, and fine-tunes the model on filtered samples.
result RAFT improves model performance in reward learning and automated metrics.
This work improves transferability of rewards inferred from expert demonstrations.
problem Transferability of rewards inferred from expert demonstrations under limited access to the expert's policy.
method Proposed principal angles as a measure of similarity and dissimilarity between transition laws. Established sufficient conditions for transferability under limited access.
result Two key results on sufficient conditions for transferability to any and local changes in transition laws.
We discuss a natural game of competition and solve the corresponding mean field game with \emph{common noise} when agents' rewards are \emph{rank dependent}. We use this solution to provide an approximate Nash equilibrium for the finite player game and obtain the rate of convergence.
Safe exploration in RF-RL doesn't increase sample complexity.
problem Achieving optimal policies with safety constraints in reward-free RL.
method Proposed SWEET framework for tabular and low-rank MDP settings, leveraging truncated value functions.
result Sample complexities match or outperform constraint-free counterparts, proving safety constraints have little impact.
New algorithms for efficient learning with long-term rewards in contextual bandits.
problem Efficient learning with long-term rewards in contextual bandits.
method Proposes new algorithms leveraging sparsity to discover dependence patterns and arm parameters.
result Regret upper bounds for data-poor and data-rich regimes, showing improved sample complexity.
A new framework assesses financial and ESG risks for sustainable investing.
problem Measuring risk and reward in sustainable investing considering environmental, social, and governance factors.
method Proposes axiomatic definitions for ESG-coherent risk measures and reward-risk ratios based on bivariate random variables.
result Empirical analysis ranks stocks using the proposed measures.
IDS improves RLHF by smoothing reward data, enhancing model performance.
problem Reward model performance degrades and overoptimization hinders true objective.
method Iterative Data Smoothing (IDS) updates model and data labels during each epoch.
result IDS outperforms traditional methods in RLHF.
Optimizes ranking from click feedback in a bandit setting.
problem Learning to rank from Bernoulli click feedback in a bandit setting.
method Variance-aware confidence sets derived from Bernstein and Chernoff bounds for optimal algorithms.
result Optimal algorithms for the case of small mean rewards, improving on previous suboptimal results.
Optimizes algorithms for non-concave bandit problems.
problem Optimizing algorithms for non-concave bandit problems.
method Unified zeroth-order optimization paradigm.
result Minimax-optimal algorithms in the dimension for low-rank generalized linear bandit problems.
Bayesian REX learns Atari games from demonstrations efficiently.
problem Bayesian reward learning for complex control problems is computationally intractable.
method Bayesian Reward Extrapolation (Bayesian REX) pre-trains a low-dimensional feature encoding and uses preferences to perform fast Bayesian inference.
result Bayesian REX learns Atari games from demonstrations in 5 minutes, competitive with state-of-the-art methods.
SARA uses similarity to learn rewards robustly and adaptively.
problem Robustness to labeler errors and adaptability to diverse feedback formats.
method Contrastive framework that learns latent representations and computes rewards as similarities.
result Strong performance on offline RL benchmarks and diverse applications.
Paper analyzes AIRL in high-dimensional spaces using random matrix theory.
problem AIRL's performance challenges in high-dimensional environments.
method Examined the rank of the matrix derived from transition matrix, applied random matrix theory.
result High-dimensional scenarios reveal transfer limitations not inherent to AIRL framework.
TLRS improves predictive power of mined formulaic alpha factors.
problem Sparse rewards in RL for mining formulaic alpha factors.
method Trajectory-level Reward Shaping (TLRS) with reward centering.
result TLRS boosts predictive power by 9.29% over existing methods.
New method extends low-rank MDPs to continuous action spaces.
problem Limited applicability of current low-rank MDP methods to continuous action spaces.
method Extending FLAMBE algorithm to continuous action spaces with Hölder smoothness conditions.
result Similar PAC bound achieved for continuous actions with polynomial dependence on smoothness order.
New algorithm tackles bilinear bandit problem with low-rank structure.
problem Finding the optimal action in a bilinear bandit problem with low-rank reward matrix.
method Two-stage algorithm: subspace exploration followed by linear bandit refinement.
result Regret bound of ESTR is O ~ ( ( d 1 + d 2 ) 3 / 2 r T ) \widetilde{\mathcal{O}}((d_1+d_2)^{3/2} \sqrt{r T}) O (( d 1 + d 2 ) 3/2 r T ) . New method for online low-rank matrix completion with improved regret.
problem Designing an efficient algorithm for online recommendation systems with low regret.
method Explore-then-commit (ETC) approach and iterative user clustering (OCTAL) for rank-1 setting.
result Nearly optimal regret bounds for online low-rank matrix completion.
Paper extends ranking metrics theory for financial positions.
problem Developing a new class of functionals for evaluating financial positions.
method Axiomatic framework based on monotonicity and cash-quasiconcavity.
result Linking ranking metrics to families of acceptance sets and risk measures.
Paper extends ranking metrics theory for financial positions.
problem Developing a new class of performance evaluation methods.
method Axiomatic framework based on monotonicity and cash-quasiconcavity.
result Linking ranking metrics to families of acceptance sets and risk measures.
This work compares human feedback methods for reward learning in bandits.
problem Understanding how human feedback affects the performance of reward learning methods.
method Theoretical comparison of human feedback approaches in offline contextual bandits.
result Human bias and uncertainty in feedback modeling impact the theoretical guarantees of reward learning methods.
Efficient algorithms for low-rank bandits using subspace recovery.
problem Contextual bandits with low-rank reward matrices.
method Spectral methods for subspace recovery, reformulating as linear bandits.
result Nearly optimal policy evaluation and best policy identification, minimax guarantees for regret minimization.
We consider the problem of active coarse ranking, where the goal is to sort items according to their means into clusters of pre-specified sizes, by adaptively sampling from their reward distributions. This setting is useful in many social science applications involving human raters and the approximate rank of every ite…
Proposes a new theoretical framework for PbRL that requires less human feedback.
problem Lack of theoretical work capturing practical PbRL frameworks.
method Introduces a reward-agnostic PbRL framework that acquires exploratory trajectories before human feedback.
result Demonstrates improved sample complexity for learning optimal policies in linear and low-rank MDPs.
Algorithm achieves comparable performance to fully dynamic data with only a few batches.
problem High-dimensional multi-armed contextual bandits with batched feedback.
method Provable sample-efficient algorithm using batch allocation method.
result Achieves regret bounds comparable to fully sequential setting with only L = O(log T) batches.
New method to rank metrics on non-shuffled traffic.
problem Position bias in ranking metrics due to item display order.
method Leverage stochasticity of recommendation policy to mitigate position bias.
result Improved ranking metrics without shuffling recommendations.
Privacy-preserving multi-party contextual bandits learn without sharing data.
problem Privacy-preserving learning for contextual bandits with multiple parties.
method Secure multi-party computation combined with epsilon-greedy differential privacy.
result Developed a privacy-preserving multi-party contextual bandit algorithm.
New estimator GMIPS reduces variance in ranking policy evaluation.
problem High variance in off-policy evaluation for ranking policies.
method GMIPS estimator with user behavior model on ranking embedding spaces.
result GMIPS achieves lowest MSE and balances bias-variance trade-off.
Study ranking in generalized linear bandits with position and item dependencies.
problem Complex reward function due to position and item dependencies in recommendation systems.
method Model position and item dependencies, design UCB and Thompson Sampling algorithms.
result Generalizes existing studies in position dependencies and graph theory.