Combining Learning from Human Feedback and Knowledge Engineering to Solve Hierarchical Tasks in Minecraft
This addresses the challenge of applying AI to poorly defined real-world tasks for researchers and practitioners in reinforcement learning and human-AI interaction, though it is incremental in combining existing techniques.
The authors tackled the problem of solving hierarchical tasks in Minecraft defined only by natural language without a reward function, by combining learning from human feedback with knowledge engineering, resulting in a solution that won first place and was awarded the most human-like agent in the 2021 NeurIPS MineRL BASALT Challenge.
Real-world tasks of interest are generally poorly defined by human-readable descriptions and have no pre-defined reward signals unless it is defined by a human designer. Conversely, data-driven algorithms are often designed to solve a specific, narrowly defined, task with performance metrics that drives the agent's learning. In this work, we present the solution that won first place and was awarded the most human-like agent in the 2021 NeurIPS Competition MineRL BASALT Challenge: Learning from Human Feedback in Minecraft, which challenged participants to use human data to solve four tasks defined only by a natural language description and no reward function. Our approach uses the available human demonstration data to train an imitation learning policy for navigation and additional human feedback to train an image classifier. These modules, combined with an estimated odometry map, become a powerful state-machine designed to utilize human knowledge in a natural hierarchical paradigm. We compare this hybrid intelligence approach to both end-to-end machine learning and pure engineered solutions, which are then judged by human evaluators. Codebase is available at https://github.com/viniciusguigo/kairos_minerl_basalt.