Michał Jamroż - Class fitting in residual convolutional networks | ML in PL 23

We leverage probabilistic models of neural representations to investigate how residual networks fit classes. To this end, we estimate class-conditional density models for representations learned by deep ResNets. We then use these models to characterize distributions of representations across learned classes. Surprisingly, we find that classes in the investigated models are not fitted in a uniform way. On the contrary: we uncover two groups of classes that are fitted with markedly different distributions of representations. These distinct modes of class-fitting are evident only in the deeper layers of the investigated models, indicating that they are not related to low-level image features. We show that the uncovered structure in neural representations correlate with memorization of training examples and adversarial robustness. Finally, we compare class-conditional distributions of neural representations between memorized and typical examples. This allows us to uncover where in the network structure class labels arise for memorized and standard inputs. The paper that concerns this work - "Neural Representations Reveal Distinct Modes of Class Fitting in Residual Convolutional Networks" was accepted at AAAI-2023 conference.
Michał Jamroż is a Data Scientist with an engineering background who has 9 years of industrial experience in Machine Learning. He is also practicing academic research as a Ph.D. candidate, where my interests revolve around deep architectures, representation learning, and probabilistic modeling.
The talk was delivered during ML in PL Conference 2023 as a part of Contributed Talks. The conference was organized by a non-profit NGO called ML in PL Association.
ML in PL Association website: mlinpl.org/
ML In PL Conference 2023 website: conference2023...
ML In PL Conference 2024 website: conference.mli...
---
ML in PL Association was founded based on the experiences in organizing of the ML in PL Conference (formerly PL in ML), the ML in PL Association is a non-profit organization devoted to fostering the machine learning community in Poland and Europe and promoting a deep understanding of ML methods. Even though ML in PL is based in Poland, it seeks to provide opportunities for international cooperation.