In-Context Learning of Energy Functions · arXivDesk
2406.12785Jun 18, 2024Proceedings of the 1st Workshop on In-Context Learning at the 41st International Conference on Machine Learning, Vienna, Austria. 2024. arXiv admin note: text overlap with arXiv:2402.10202
In-context learning is a powerful capability of certain machine learning models that arguably underpins the success of today's frontier AI models. However, in-context learning is critically limited to settings where the in-context distribution of interest pθICL(x∣D) can be straightforwardly expressed and/or parameterized by the model; for instance, language modeling relies on expressing the next-token distribution as a categorical distribution parameterized by the network's output logits. In this work, we present a more general form of in-context learning without such a limitation that we call in-context learning of energy functions. The idea is to instead learn the unconstrained and arbitrary in-context energy function EθICL(x∣D)
Nearby in the stack
corresponding to the in-context distribution
pθICL(x∣D)
. To do this, we use classic ideas from energy-based modeling. We provide preliminary evidence that our method empirically works on synthetic data. Interestingly, our work contributes (to the best of our knowledge) the first example of in-context learning where the input space and output space differ from one another, suggesting that in-context learning is a more-general capability than previously realized.