Human-Guided Complexity-Controlled Abstractions
Keywords: human-in-the-loop, representation learning, interpretability
TL;DR: Generating good abstractions enables rapid adaptation, but identifying such abstractios a priori is challenging; a human-in-the-loop framework allows people to select the right abstraction level in models.
Abstract: Neural networks often learn task-specific latent representations that fail to generalize to novel settings or tasks. Conversely, humans learn discrete representations (i.e., concepts or words) at a variety of abstraction levels (e.g., "bird" vs. "sparrow'") and use the appropriate abstraction based on tasks. Inspired by this, we train neural models to generate a spectrum of discrete representations, and control the complexity of the representations (roughly, how many bits are allocated for encoding inputs) by tuning the entropy of the distribution over representations. In finetuning experiments, using only a small number of labeled examples for a new task, we show that (1) tuning the representation to a task-appropriate complexity level supports the greatest finetuning performance, and (2) in a human-participant study, users were able to identify the appropriate complexity level for a downstream task via visualizations of discrete representations. Our results indicate a promising direction for rapid model finetuning by leveraging human insight.
Supplementary Material: pdf
Submission Number: 12633