On the Complexity of Neural Computation in Superposition
Recent advances in the understanding of neural networks suggest that superposition, the ability of a single neuron to represent multiple features simultaneously, is a key mechanism underlying the computational efficiency of large-scale networks. This paper explores the theoretical foundations of com...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | en_US |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/1721.1/157073 |