Ex-Model: Continual Learning from a Stream of Trained Models
This addresses the problem of inefficient, non-scalable, and privacy-limited continual learning for AI systems by shifting from raw data to model-based learning, though it appears incremental as it builds on existing continual learning concepts.
The paper tackles the challenge of continual learning from non-stationary data streams by proposing a new paradigm called Ex-Model Continual Learning (ExML), where agents learn from sequences of trained models instead of raw data, and demonstrates its effectiveness through three algorithms tested on datasets like MNIST, CIFAR-10, and CORe50 across eight scenarios.
Learning continually from non-stationary data streams is a challenging research topic of growing popularity in the last few years. Being able to learn, adapt, and generalize continually in an efficient, effective, and scalable way is fundamental for a sustainable development of Artificial Intelligent systems. However, an agent-centric view of continual learning requires learning directly from raw data, which limits the interaction between independent agents, the efficiency, and the privacy of current approaches. Instead, we argue that continual learning systems should exploit the availability of compressed information in the form of trained models. In this paper, we introduce and formalize a new paradigm named "Ex-Model Continual Learning" (ExML), where an agent learns from a sequence of previously trained models instead of raw data. We further contribute with three ex-model continual learning algorithms and an empirical setting comprising three datasets (MNIST, CIFAR-10 and CORe50), and eight scenarios, where the proposed algorithms are extensively tested. Finally, we highlight the peculiarities of the ex-model paradigm and we point out interesting future research directions.