In-Context Learning of Energy Functions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Schaeffer, Rylan, Khona, Mikail, Koyejo, Sanmi
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911924457832448
author Schaeffer, Rylan
Khona, Mikail
Koyejo, Sanmi
author_facet Schaeffer, Rylan
Khona, Mikail
Koyejo, Sanmi
contents 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 $p_θ^{ICL}( x|\mathcal{D})$ 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 \textit{in-context learning of energy functions}. The idea is to instead learn the unconstrained and arbitrary in-context energy function $E_θ^{ICL}(x|\mathcal{D})$ corresponding to the in-context distribution $p_θ^{ICL}(x|\mathcal{D})$. 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.
format Preprint
id arxiv_https___arxiv_org_abs_2406_12785
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle In-Context Learning of Energy Functions
Schaeffer, Rylan
Khona, Mikail
Koyejo, Sanmi
Machine Learning
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 $p_θ^{ICL}( x|\mathcal{D})$ 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 \textit{in-context learning of energy functions}. The idea is to instead learn the unconstrained and arbitrary in-context energy function $E_θ^{ICL}(x|\mathcal{D})$ corresponding to the in-context distribution $p_θ^{ICL}(x|\mathcal{D})$. 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.
title In-Context Learning of Energy Functions
topic Machine Learning
url https://arxiv.org/abs/2406.12785