Vivian White

Vivian White

she/her

CS Master's student interested in computational neuroscience and machine learning

Publications

You can also find my articles on my Google Scholar profile.

Workshop Papers

Learning Stochastic Rainbow Networks

Published in NeurIPS Sci4DL Workshop, 2024

Random feature models are a popular approach for studying network learning that can capture important behaviors while remaining simpler than traditional training. Guth et al. [2024] introduced “rainbow” networks which model the distribution of trained weights as correlated random features conditioned on previous layer activity. Sampling new weights from distributions fit to learned networks led to similar performance in entirely untrained networks, and the observed weight covariance were found to be low rank. This provided evidence that random feature models could be extended to some networks away from initialization. However, White et al. [2024] failed to replicate their results in the deeper ResNet18 architecture. Here we ask whether the rainbow formulation can succeed in deeper networks by directly training a stochastic ensemble of random features, which we call stochastic rainbow networks. At every gradient descent iteration, new weights are sampled for all intermediate layers and features aligned layer-wise. We find: (1) this approach scales to deeper models, which outperform shallow networks at large widths; (2) ensembling multiple samples from the stochastic model is better than retraining the classifier head; and (3) low-rank parameterization of the learnable weight covariances can approach the accuracy of full-rank networks. This offers more evidence for rainbow and other structured random feature networks as reduced models of deep learning.

Recommended citation: Vivian White, Muawiz Sajjad Chaudhary, Guy Wolf, Guillaume Lajoie, and Kameron Decker Harris. Learning stochastic rainbow networks. In NeurIPS 2024 Workshop on Scientific Methods for Understanding Deep Learning, 2024. URL https://openreview.net/pdf?id=bvJ4UGTRtk.
Download Paper

Learning and Aligning Structured Random Feature Networks

Published in ICLR Re-Align Workshop, 2024

Artificial neural networks (ANNs) are considered “black boxes” due to the difficulty of interpreting their learned weights. While choosing the best features is not well understood, random feature networks (RFNs) and wavelet scattering ground some ANN learning mechanisms in function space with tractable mathematics. Meanwhile, the genetic code has evolved over millions of years, shaping the brain to develop variable neural circuits with reliable structure that resemble RFNs. We explore a similar approach, embedding neuro-inspired, wavelet-like weights into multilayer RFNs. These can outperform scattering and have kernels that describe their function space at large width. We build learnable and deeper versions of these models where we can optimize separate spatial and channel covariances of the convolutional weight distributions. We find that these networks can perform comparatively with conventional ANNs while dramatically reducing the number of trainable parameters. Channel covariances are most influential, and both weight and activation alignment are needed for classification performance. Our work out lines how neuro-inspired configurations may lead to better performance in key cases and offers a potentially tractable reduced model for ANN learning.

Recommended citation: Vivian White, Muawiz Sajjad Chaudhary, Guy Wolf, Guillaume Lajoie, and Kameron Decker Harris. Learning and aligning structured random feature networks. In ICLR 2024 Workshop on Representational Alignment, 2024. URL https://openreview.net/forum?id=vWhUQXQoFF.
Download Paper