TOPPO: Rethinking PPO for Multi-Task Reinforcement Learning with Critic Balancing
cs.AI, cs.LG, cs.RO, stat.ML
Submitted: 2026-05-12
Updated: 2026-11-07
Code: https://github.com/ms-cho/Adaptive-Reward-Scaling-ARS-for-MTRL
Terminology
Sources
- What Matters In On-Policy Reinforcement Learning? A Large-Scale Empirical Study
- Layer Normalization
- Fair Resource Allocation in Multi-Task Learning
- Implementation Matters in Deep Policy Gradients: A Case Study on PPO and TRPO
- TD-MPC2: Scalable, Robust World Models for Continuous Control
- Multi-Task Reinforcement Learning with Mixture of Orthogonal Experts
- Multi-task Deep Reinforcement Learning with PopArt
- Benchmarking Massively Parallelized Multi-Task Reinforcement Learning for Robotics Tasks
- SimBa: Simplicity Bias for Scaling Up Parameters in Deep Reinforcement Learning
- FAMO: Fast Adaptive Multitask Optimization
- Meta-World+: An Improved, Standardized, RL Benchmark
- High-Dimensional Continuous Control Using Generalized Advantage Estimation
- Proximal Policy Optimization Algorithms
- Multi-Task Reinforcement Learning with Context-based Representations
- PaCo: Parameter-Compositional Multi-Task Reinforcement Learning
- Learning values across many orders of magnitude
- Multi-Task Reinforcement Learning with Soft Modularization
- Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents