IQ-Learn: Inverse soft-Q Learning for Imitation

In many sequential decision-making problems (e.g., robotics control, game playing, sequential prediction), human or expert data is available containing useful information about the task. However, imitation learning (IL) from a small amount of expert data can be challenging in high-dimensional environments with complex dynamics. Behavioral cloning is a simple method that is widely used due to its simplicity of implementation and stable convergence but doesn't utilize any information involving the environment's dynamics. Many existing methods that exploit dynamics information are difficult to train in practice due to an adversarial optimization process over reward and policy approximators or biased, high variance gradient estimators. We introduce a method for dynamics-aware IL which avoids adversarial training by learning a single Q-function, implicitly representing both reward and policy. On standard benchmarks, the implicitly learned rewards show a high positive correlation with the ground-truth rewards, illustrating our method can also be used for inverse reinforcement learning (IRL). Our method, Inverse soft-Q learning (IQ-Learn) obtains state-of-the-art results in offline and online imitation learning settings, significantly outperforming existing methods both in the number of required environment interactions and scalability in high-dimensional spaces, often by more than 3x.

PDF Abstract NeurIPS 2021 PDF NeurIPS 2021 Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
MuJoCo Games Ant IQ-Learn Average Return 4362.9 # 1
Atari Games Atari 2600 Beam Rider IQ-Learn Return 3025 # 1
Atari Games Atari 2600 Q*Bert IQ-Learn Return 12940 # 1
Atari Games Atari 2600 Seaquest IQ-Learn Return 2349 # 1
Atari Games Atari 2600 Space Invaders IQ-Learn Return 507 # 1
MuJoCo Games Humanoid-v2 IQ-Learn Return 5227.1 # 1
MuJoCo Games Walker2d IQ-Learn Mean 5134 # 1

Methods