Coreset-Based Task Selection for Sample-Efficient Meta-Reinforcement Learning
math.OC, cs.LG
Submitted: 2025-02-04
Updated: 2026-09-26
Comments: Paper accepted at CDC 2025
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- A Moreau Envelope Approach for LQR Meta-Policy Estimation
- RL$^2$: Fast Reinforcement Learning via Slow Reinforcement Learning
- Online convex optimization in the bandit setting: gradient descent without a gradient
- Regret Analysis of Multi-task Representation Learning for Linear-Quadratic Adaptive Control
- On the Convergence of FedAvg on Non-IID Data
- Convergence of Gradient-based MAML in LQR
- Learning to Adapt in Dynamic, Real-World Environments Through Meta-Reinforcement Learning
- Model-Agnostic Zeroth-Order Policy Optimization for Meta-Learning of Ergodic Linear Quadratic Regulators
- ProMP: Proximal Meta-Policy Search
- Evolution Strategies as a Scalable Alternative to Reinforcement Learning
- On Task-Relevant Loss Functions in Meta-Reinforcement Learning and Online LQR
- ES-MAML: Simple Hessian-Free Meta Learning
- Local SGD Converges Fast and Communicates Little
- Asynchronous Heterogeneous Linear Quadratic Regulator Design
- Gymnasium: A Standard Interface for Reinforcement Learning Environments
- Model-free Learning with Heterogeneous Dynamical Systems: A Federated LQR Approach
- Learning to reinforcement learn
- Towards Sustainable Learning: Coresets for Data-efficient Deep Learning
- Preparing for the Unknown: Learning a Universal Policy with Online System Identification
Related papers
- Lions and Muons: Optimization via Stochastic Frank-Wolfe under Heavy-Tailed Noise
- Adam-HNAG: A Convergent Reformulation of Adam with Accelerated Rate
- Incremental Learning in Mirror Flows
- Second-Order Optimization with Asynchronous Curvature
- Online Control via Counterfactual Tracking
- Asynchronous Replanning in Two Population Linear Quadratic Mean Field Games: Information Requirements and Stability