Multi-Task RL

RL: Reinforcement Learning

Overview

Multi-task reinforcement learning studies how an agent can learn several related objectives together, reusing experience to reduce data needs and transfer skills to new tasks. Common approaches include task-conditioned policies, shared representations and world models, meta-learning, successor features, and transformer or mixture-of-experts policies; curricula and distillation help allocate learning across tasks. Methods are tested in settings such as Meta-World manipulation, multi-agent benchmarks, robotics, network control, and multi-domain language-model training. A central challenge is balancing useful sharing against negative transfer: a model may improve on average while weakening on particular tasks or failing to generalize to unseen ones.

Latest papers

92 papers7 papers in the last four weeks, down 12% on the four weeks before. 0.1% of all new papers.

Jun 19, 2025

Energy-Based Transfer for Reinforcement Learning

Reinforcement learning remains sample inefficient, particularly when agents must adapt over time to related but distinct tasks, as in continual learning settings. Efficiency can be improved by transferring knowledge from a pretrained teacher policy to guide exploration in new but related tasks. However, if the new task sufficiently differs from the teacher's training task, the transferred guidance may be sub-optimal and bias exploration toward low-reward behaviors. We propose an energy-based transfer learning method that uses out-of-distribution detection to selectively issue guidance, enabling the teacher to intervene only in states within its training distribution. We theoretically link energy scores to state-visitation patterns under on-policy reinforcement learning. Empirically, our method improves sample efficiency and performance across both single-task and multi-task settings.
Dec 6, 2024

Enabling Energy-Efficient Simultaneous Multi-Task Reinforcement Learning through Spiking Neural Networks with Active Dendrites for Bio-inspired Generalist Agents

Reinforcement learning (RL) has demonstrated remarkable capabilities in training agents to solve complex tasks autonomously, such as mobile robots, UAVs/UGVs, and game-playing agents). However, scaling RL to master multiple tasks simultaneously (i.e., so-called multi-task RL) remains a significant challenge. Such a multi-task RL capability especially is important for agents to adapt to changes in real-world operational environments. State-of-the-art works show that, training agents with neural networks and shared structures across tasks promises improved generalization in simultaneous multi-task RL. However, they still suffer from task interference and incur high energy consumption due to intensive computation. To address this, we propose MTSpark, a novel methodology that enables energy-efficient simultaneous multi-task RL using spiking neural networks (SNNs) equipped with active dendrites for bio-inspired generalist agents. Specifically, MTSpark enhances a Deep Spiking Q-Network (DSQN) with active dendrites, a dueling structure, and task-specific context signals to dynamically form specialized sub-networks for individual tasks, while exploiting sparse operations for energy-efficient network processing. Experimental results demonstrate that MTSpark achieves higher performance and efficiency compared to state-of-the-art by obtaining high scores across three Atari games (i.e., Pong: -5.4, Breakout: 0.6, and Enduro: 371.2), approaching human-level performance (i.e., Pong: -3, Breakout: 31, Enduro: 368), while incurring similar memory and about 2x lower energy than state-of-the-art. These results show that our MTSpark potentially advances the frontiers toward energy-efficient generalist agents by combining RL and SNNs.