# On the special role of class-selective neurons in early training

Source: [On the special role of class-selective neurons in early training](<https://www.sota2.com/research/paper/68fa6f781e209131cea45cac>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T01:20:27.449Z

Authors: Omkar Ranadive, Nikhil Thakurdesai, Ari S Morcos, Matthew Leavitt, St\\'ephane Deny

Published: 2023-05-27T08:22:34.000Z

## Abstract

It is commonly observed that deep networks trained for classification exhibit class-selective neurons in their early and intermediate layers. Intriguingly, recent studies have shown that these class-selective neurons can be ablated without deteriorating network function. But if class-selective neurons are not necessary, why do they exist? We attempt to answer this question in a series of experiments on ResNet-50s trained on ImageNet. We first show that class-selective neurons emerge during the first few epochs of training, before receding rapidly but not completely; this suggests that class-selective neurons found in trained networks are in fact vestigial remains of early training. With single-neuron ablation experiments, we then show that class-selective neurons are important for network function in this early phase of training. We also observe that the network is close to a linear regime in this early phase; we thus speculate that class-selective neurons appear early in training as quasi-linear shortcut solutions to the classification task. Finally, in causal experiments where we regularize against class selectivity at different points in training, we show that the presence of class-selective neurons early in training is critical to the successful training of the network; in contrast, class-selective neurons can be suppressed later in training with little effect on final accuracy. It remains to be understood by which mechanism the presence of class-selective neurons in the early phase of training contributes to the successful training of networks.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2305.17409>)

[PDF](<https://arxiv.org/pdf/2305.17409.pdf>)
