Contrastive Similarity Matching for Supervised Learning

Abstract

We propose a novel biologically-plausible solution to the credit assignment problem motivated by observations in the ventral visual pathway and trained deep neural networks. In both, representations of objects in the same category become progressively more similar, while objects belonging to different categories become less similar. We use this observation to motivate a layer-specific learning goal in a deep network: each layer aims to learn a representational similarity matrix that interpolates between previous and later layers. We formulate this idea using a contrastive similarity matching objective function and derive from it deep neural networks with feedforward, lateral, and feedback connections, and neurons that exhibit biologically-plausible Hebbian and anti-Hebbian plasticity. Contrastive similarity matching can be interpreted as an energy-based learning algorithm, but with significant differences from others in how a contrastive function is constructed.

References in corpus (8)

• Direct Feedback Alignment Provides Learning in Deep Neural Networks
• Why do similarity matching objectives lead to Hebbian/anti-Hebbian networks?
• Blind nonnegative source separation using biological neural networks
• A Normative Theory of Adaptive Dimensionality Reduction in Neural Networks
• STDP as presynaptic activity times rate of change of postsynaptic activity
• Greedy Layerwise Learning Can Scale to ImageNet
• Conducting Credit Assignment by Aligning Local Representations
• Learning From Brains How to Regularize Machines

Cited by in corpus (1)

• Kernelized information bottleneck leads to biologically plausible 3-factor Hebbian learning in deep networks