Abstract
We extend feed-forward neural networks with a Dirichlet process prior over the weight distribution. This enforces a sharing on the network weights, which can reduce the overall number of parameters drastically. We alternately sample from the posterior of the weights and the posterior of assignments of network connections to the weights. This results in a weight sharing that is adopted to the given data. In order to make the procedure feasible, we present several techniques to reduce the computational burden. Experiments show that our approach mostly outperforms models with random weight sharing. Our model is capable of reducing the memory footprint substantially while maintaining a good performance compared to neural networks without weight sharing.
Original language | English |
---|---|
Pages (from-to) | 246-252 |
Journal | IEEE Transactions on Pattern Analysis and Machine Intelligence |
Volume | 42 |
Issue number | 1 |
DOIs | |
Publication status | Published - 2020 |
Keywords
- Bayesian neural networks
- Dirichlet processes
- Gibbs sampling
- hybrid Monte-Carlo
- non-conjugate models
- weight sharing
ASJC Scopus subject areas
- Software
- Computer Vision and Pattern Recognition
- Computational Theory and Mathematics
- Artificial Intelligence
- Applied Mathematics