Aldo Pacchiano

Latest

• State-free Reinforcement Learning
• Second Order Bounds for Contextual Bandits with Function Approximation
• Learning Rate-Free Reinforcement Learning: A Case for Model Selection with Non-Stationary Objectives
• Provable Interactive Learning with Hindsight Instruction Feedback
• Multiple-policy Evaluation via Density Estimation
• A Framework for Partially Observed Reward-States in RLHF
• Provably Sample Efficient RLHF via Active Preference Optimization
• Data-Driven Regret Balancing for Online Model Selection in Bandits
• Contextual Bandits with Stage-wise Constraints
• Improving Offline RL by Blending Heuristics
• Experiment Planning with Function Approximation
• Anytime Model Selection in Linear Bandits
• Supervised Pretraining Can Learn In-Context Reinforcement Learning
• Transfer RL via the Undo Maps Formalism
• A Unified Model and Dimension for Interactive Estimation
• Leveraging Offline Data in Online Reinforcement Learning
• Estimating Optimal Policy Value in General Linear Contextual Bandits
• Parallelizing Contextual Bandits
• Dueling RL: Reinforcement Learning with Trajectory Preferences
• Neural Design for Genetic Perturbation Experiments
• An Instance-Dependent Analysis for the Cooperative Multi-Player Multi-Armed Bandit
• Learning General World Models in a Handful of Reward-Free Deployments
• Unpacking Reward Shaping: Understanding the Benefits of Reward Engineering on Sample Complexity
• Best of Both Worlds Model Selection
• Joint Representation Training in Sequential Tasks with Shared Structure
• Online Nonsubmodular Minimization with Delayed Costs: From Full Information to Bandit Feedback
• Meta Learning MDPs with Linear Transition Models
• ES-ENAS: Blackbox Optimization over Hybrid Spaces via Combinatorial and Continuous Evolution
• Neural Pseudo-Label Optimism for the Bank Loan Problem
• Towards an Understanding of Default Policies in Multitask Policy Optimization
• Reinforcement Learning in Linear MDPs: Constant Regret and Representation Selection
• Unlocking Pixels for Reinforcement Learning via Implicit Attention
• Towards Tractable Optimism in Model-Based Reinforcement Learning
• Dynamic Balancing for Model Selection in Bandits and RL
• Sample Efficient Reinforcement Learning In Continuous State Spaces: A Perspective Beyond Linearity
• Model Selection for Contextual Bandits and Reinforcement Learning
• On the Theory of Reinforcement Learning with Once-per-Episode Feedback
• Parallelizing Contextual Linear Bandits
• Learning the Truth From Only One Side of the Story
• Stochastic Bandits with Linear Constraints
• Near Optimal Policy Optimization via REPS
• Tactical Optimism and Pessimism for Deep Reinforcement Learning
• Fairness with Continuous Optimal Transport
• Effective Diversity in Population-Based Reinforcement Learning
• Model Selection in Contextual Stochastic Bandit Problems
• Ridge Rider: Finding Diverse Solutions by Following Eigenvectors of the Hessian
• Accelerated Message Passing for Entropy-Regularized MAP Inference
• Learning to Score Behaviors for Guided Policy Optimization
• On Approximate Thompson Sampling with Langevin Algorithms
• Ready Policy one: World Building Through Active Learning
• Stochastic Flows and Geometric Optimization on the Orthogonal Group
• Online Model Selection for Reinforcement Learning with Function Approximation
• Regret Bound Balancing and Elimination for Model Selection in Bandits and RL
• Wasserstein Fair Classification
• Taming the Herd: Multi-Modal Meta-Learning with a Population of Agents
• Regret Balancing for Bandit and RL Model Selection
• Convergence Rates of Smooth Message Passing with Rounding in Entropy-Regularized MAP Inference
• Practical Nonisotropic Monte Carlo Sampling in High Dimensions via Determinantal Point Processes
• Provably Robust Blackbox Optimization for Reinforcement Learning
• Robustness Guarantees for Mode Estimation with an Application to Bandits
• From Complexity to Simplicity: Adaptive ES-Active Subspaces for Blackbox Optimization
• ES-MAML: Simple Hessian-Free Meta Learning
• Reinforcement Learning with Chromatic Networks for Compact Architecture Search
• Computing Stable Solutions in Threshold Network Flow Games With Bounded Treewidth
• KAMA-NNs: Low-Dimensional Rotation Based Neural Networks
• Gen-Oja: A Two-time-scale approach for Streaming CCA
• Online learning with kernel losses
• Reinforcement Learning with Wasserstein Distance Regularisation, with Applications to Multipolicy Learning
• Conditions Beyond Treewidth for Tightness of Higher-order LP Relaxations
• Real Time Clustering of Time Series Using Triangular Potentials
• Computational Approaches to Poisson Traces Associated to Finite Subgroups of Sp2n(C)
• A General Approach to Fairness with Optimal Transport
• Geometrically Coupled Monte Carlo Sampling
• Trace Reconstruction Problem