Two Studies of Constraints in High Dimensions: Entropy Inequalities and the Randomized Symmetric Binary Perceptron
We study two constrained problems in high dimensions. We study a high dimensional inequality for the binary entropy. The perceptron is a natural model in high-dimensional probability, and a toy shallow neural network which stores random patterns; we also study a randomized variant of the symmetric b...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Massachusetts Institute of Technology
2024
|
| Online Access: | https://hdl.handle.net/1721.1/153999 |