Task-Relevant Adversarial Imitation Learning

Zolna Konrad

[0]

Reed Scott

[0]

Novikov Alexander

[0]

Colmenarej Sergio Gomez

Budden David

[0]

Cabi Serkan

Denil Misha

[0]

de Freitas Nando

[0]

Wang Ziyu

[0]

Cited by: 11|Views100

Keywords:

D4PG from DemonstrationsGenerative Adversarial NetworksMarkov Decision Processinverse reinforcement learningdeterministic policyMore(11+)

Weibo:

We showed through ablations that Task-Relevant Adversarial Imitation Learning benefits from data augmentation, actor early stopping, and the use of invariant set constraints

Abstract:

We show that a critical problem in adversarial imitation from high-dimensional sensory data is the tendency of discriminator networks to distinguish agent and expert behaviour using task-irrelevant features beyond the control of the agent. We analyze this problem in detail and propose a solution as well as several baselines that outperf...More

Code:

Data:

Full Text

View via Publisher