Abstract
Not being able to understand and predict the behavior of deep learning systems makes it hard to decide what architecture and algorithm to use for a given problem. In science and engineering, modeling is a methodology used to understand complex systems whose internal processes are opaque. Modeling replaces a complex system with a simpler, more interpretable surrogate. Drawing inspiration from this, we construct a class of surrogate models for neural networks using Gaussian processes. Rather than deriving kernels for infinite neural networks, we learn kernels empirically from the naturalistic behavior of finite neural networks. We demonstrate our approach captures existing phenomena related to the spectral bias of neural networks, and then show that our surrogate models can be used to solve practical problems such as identifying which points most influence the behavior of specific neural networks and predicting which architectures and algorithms will generalize well for specific datasets.
Original language | English (US) |
---|---|
Pages (from-to) | 1241-1252 |
Number of pages | 12 |
Journal | Proceedings of Machine Learning Research |
Volume | 216 |
State | Published - 2023 |
Externally published | Yes |
Event | 39th Conference on Uncertainty in Artificial Intelligence, UAI 2023 - Pittsburgh, United States Duration: Jul 31 2023 → Aug 4 2023 |
All Science Journal Classification (ASJC) codes
- Artificial Intelligence
- Software
- Control and Systems Engineering
- Statistics and Probability