Compete and Compose: Learning Independent Mechanisms for Modular World Models
We present COmpetitive Mechanisms for Efficient Transfer (COMET), a modular world model which leverages reusable, independent mechanisms across different environments. COMET is trained on multiple environments with varying dynamics via a two-step process: competition and composition. This enables the model to recognise and learn transferable mechanisms. Specifically, in the competition phase, COMET is trained with a winner-takes-all gradient allocation, encouraging the emergence of independent mechanisms. These are then re-used in the composition phase, where COMET learns to re-compose learnt mechanisms in ways that capture the dynamics of intervened environments. In so doing, COMET explicitly reuses prior knowledge, enabling efficient and interpretable adaptation. We evaluate COMET on environments with image-based observations. In contrast to competitive baselines, we demonstrate that COMET captures recognisable mechanisms without supervision. Moreover, we show that COMET is able to adapt to new environments with varying numbers of objects with improved sample efficiency compared to more conventional finetuning approaches.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Towards Error Centric Intelligence I, Beyond Observational Learning
We argue that progress toward AGI is theory limited rather than data or scale limited. Building on the critical rationalism of Popper and Deutsch, we challenge the Platonic Representation Hypothesis. Observationally equi…
Learning Causal Mechanisms through Orthogonal Neural Networks
A fundamental feature of human intelligence is the ability to infer high-level abstractions from low-level sensory data. An essential component of such inference is the ability to discover modularized generative mechanis…
Few-Shot and Continual Learning with Attentive Independent Mechanisms
Deep neural networks (DNNs) are known to perform well when deployed to test distributions that shares high similarity with the training distribution. Feeding DNNs with new data sequentially that were unseen in the traini…
Continual LearningFew-Shot LearningMixture-of-ExpertsModularRSI: Modular and Generalizable Recursive Harness Self-Improvement
Recent work extends recursive self-improvement (RSI) to agent harnesses for long-horizon coding and terminal tasks, enabling agents to improve execution mechanisms from experience. However, generalizable harness RSI rema…
Dissociating language and thought in large language models
Large Language Models (LLMs) have come closest among all models to date to mastering human language, yet opinions about their linguistic and cognitive capabilities remain split. Here, we evaluate LLMs using a distinction…