Is Imitation All You Need? Generalized Decision-Making with Dual-Phase Training

TL;DR

DualMind introduces a dual-phase training approach for generalist decision-making agents, enabling zero-shot task generalization across diverse domains without task-specific fine-tuning.

Contribution

The paper proposes a novel dual-phase training strategy that improves generalization and reduces overfitting in decision-making agents across multiple domains.

Findings

Outperforms previous generalist agents by over 50% on Habitat and 70% on MetaWorld.

Successfully completes over 30 MetaWorld tasks at 90% success rate.

Demonstrates zero-shot generalization across diverse tasks and environments.

Abstract

We introduce DualMind, a generalist agent designed to tackle various decision-making tasks that addresses challenges posed by current methods, such as overfitting behaviors and dependence on task-specific fine-tuning. DualMind uses a novel "Dual-phase" training strategy that emulates how humans learn to act in the world. The model first learns fundamental common knowledge through a self-supervised objective tailored for control tasks and then learns how to make decisions based on different contexts through imitating behaviors conditioned on given prompts. DualMind can handle tasks across domains, scenes, and embodiments using just a single set of model weights and can execute zero-shot prompting without requiring task-specific fine-tuning. We evaluate DualMind on MetaWorld and Habitat through extensive experiments and demonstrate its superior generalizability compared to previous techniques, outperforming other generalist agents by over 50$\%$ and 70$\%$ on Habitat and MetaWorld, respectively. On the 45 tasks in MetaWorld, DualMind achieves over 30 tasks at a 90$\%$ success rate.