Accuracy-based Curriculum Learning in Deep Reinforcement Learning

Abstract

In this paper, we investigate a new form of automated curriculum learning based on adaptive selection of accuracy requirements, called accuracy-based curriculum learning. Using a reinforcement learning agent based on the Deep Deterministic Policy Gradient algorithm and addressing the Reacher environment, we first show that an agent trained with various accuracy requirements sampled randomly learns more efficiently than when asked to be very accurate at all times. Then we show that adaptive selection of accuracy requirements, based on a local measure of competence progress, automatically generates a curriculum where difficulty progressively increases, resulting in a better learning efficiency than sampling randomly.

References in corpus (9)

• Continuous control with deep reinforcement learning
• Active Learning of Inverse Models with Intrinsically Motivated Goal Exploration in Robots
• Reproducibility of Benchmarked Deep Reinforcement Learning Tasks for Continuous Control
• Intrinsically Motivated Goal Exploration Processes with Automatic Curriculum Learning
• Curriculum Learning by Transfer Learning: Theory and Experiments with Deep Networks
• Reverse Curriculum Generation for Reinforcement Learning
• Multi-Armed Bandits for Intelligent Tutoring Systems
• Unsupervised Learning of Goal Spaces for Intrinsically Motivated Goal Exploration
• Active choice of teachers, learning strategies and goals for a socially guided intrinsic motivation learner

Cited by in corpus (5)

• Accelerating Reinforcement Learning for Reaching using Continuous Curriculum Learning
• Modular Design Patterns for Hybrid Learning and Reasoning Systems: a taxonomy, patterns and use cases
• Curiosity-Driven Multi-Criteria Hindsight Experience Replay
• Meta Automatic Curriculum Learning
• Mastering Rate based Curriculum Learning