Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces LA-MAML, a novel approach to Model-Agnostic Meta-Learning that replaces traditional trajectory collection and gradient updates in the inner loop with a learned embedding of task-specific language instructions. By leveraging these instructions, LA-MAML allows for a single-step adaptation of global policy parameters, significantly reducing the computational cost associated with inner loop updates. Experiments on the BabyAI benchmark show that LA-MAML not only maintains competitive performance but also improves training efficiency compared to existing methods.
Language instructions can replace costly trajectory-based updates in meta-reinforcement learning, leading to faster training without sacrificing performance.
Model-Agnostic Meta-Learning (MAML) is a widely used framework for reinforcement learning (RL) that enables efficient transfer by learning global policy parameters that can be rapidly adapted to new tasks. MAML training proceeds in two loops: an inner loop where the global parameters are adapted to task-specific parameters, and an outer loop where these task-specific parameters are evaluated and losses are back-propagated to improve the global parameters. Traditionally, the inner loop adaptation is performed by collecting trajectories from the task environment and applying gradient updates on the empirical expected return, which can be a costly operation. We note that it is the outer loop that drives the actual learning of global parameters, and therefore the inner loop adaptation mechanism need not be restricted to be gradient-based. This observation leads us to ask: Can we replace the inner loop trajectory collection and gradient update with a simpler, task-specific signal? In many practical settings, tasks are naturally accompanied by language instructions. Leveraging these instructions as a direct task-specific signal, we propose LA-MAML (Language Adapted MAML), which modifies the inner loop by adapting the global policy parameters in a single step through a learned embedding of the task instruction, replacing the inner loop trajectory collection and gradient-based updates. Experiments on the BabyAI benchmark demonstrate that LA-MAML achieves competitive or improved performance compared to baselines at a significantly lower per-iteration wall-clock training time. These results demonstrate that language instructions are an effective and efficient substitute for trajectory-based inner loop adaptation in meta RL.