Learning With Amigo: Adversarially Motivated Intrinsic Goals

Abstract

A key challenge for reinforcement learning (RL) consists of learning in environments with sparse extrinsic rewards. In contrast to current RL methods, humans are able to learn new skills with little or no reward by using various forms of intrinsic motivation. We propose AMIGo, a novel agent incorporating -- as form of meta-learning -- a goal-generating teacher that proposes Adversarially Motivated Intrinsic Goals to train a goal-conditioned "student" policy in the absence of (or alongside) environment reward. Specifically, through a simple but effective "constructively adversarial" objective, the teacher learns to propose increasingly challenging -- yet achievable -- goals that allow the student to learn general skills for acting in a new environment, independent of the task to be solved. We show that our method generates a natural curriculum of self-proposed goals which ultimately allows the agent to solve challenging procedurally-generated tasks where other forms of intrinsic motivat

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keycampero2020learning

Related papers

• Autotelic Reinforcement Learning: Exploring Intrinsic Motivations For Skill Acquisition In Open-ended Environments (2025)5.24
• Unsupervised Learning Of Efficient Exploration: Pre-training Adaptive Policies Via Self-imposed Goals (2026)0.00
• Autotelic Agents With Intrinsically Motivated Goal-conditioned Reinforcement Learning: A Short Survey (2020)0.00
• On Learning Intrinsic Rewards For Policy Gradient Methods (2018)0.00
• Autotelic Reinforcement Learning In Multi-agent Environments (2022)0.00
• Generating Automatic Curricula Via Self-supervised Active Domain Randomization (2020)0.00
• TGRL: An Algorithm For Teacher Guided Reinforcement Learning (2023)0.00
• LIGS: Learnable Intrinsic-reward Generation Selection For Multi-agent Learning (2021)0.00