# On robustness of generative representations against catastrophic forgetting

arXiv: https://arxiv.org/abs/2109.01844 · HTML: https://mlsanity.com/paper/2109.01844/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 31 references that are in the map.
- First posted on arXiv 4 Sep 2021.

## Builds on (most cited first, 24 of 24)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [Continual Lifelong Learning with Neural Networks: A Review](https://mlsanity.com/paper/1802.07569/) (2018)
- [Continual Learning with Deep Generative Replay](https://mlsanity.com/paper/1705.08690/) (2017)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning](https://mlsanity.com/paper/1711.05769/) (2017)
- [Lifelong Learning with Dynamically Expandable Networks](https://mlsanity.com/paper/1708.01547/) (2017)
- [Online Continual Learning with Maximally Interfered Retrieval](https://mlsanity.com/paper/1908.04742/) (2019)
- [Continual learning with hypernetworks](https://mlsanity.com/paper/1906.00695/) (2019)
- [Supermasks in Superposition](https://mlsanity.com/paper/2006.14769/) (2020)
- [Reinforced Continual Learning](https://mlsanity.com/paper/1805.12369/) (2018)
- [Anatomy of Catastrophic Forgetting: Hidden Representations and Task Semantics](https://mlsanity.com/paper/2007.07400/) (2020)
- [Alleviating catastrophic forgetting using context-dependent gating and synaptic stabilization](https://mlsanity.com/paper/1802.01569/) (2018)
- [Generative replay with feedback connections as a general strategy for continual learning](https://mlsanity.com/paper/1809.10635/) (2018)
- [Continual Learning via Neural Pruning](https://mlsanity.com/paper/1903.04476/) (2019)
- [Rehearsal revealed: The limits and merits of revisiting samples in continual learning](https://mlsanity.com/paper/2104.07446/) (2021)
- [Toward Understanding Catastrophic Forgetting in Continual Learning](https://mlsanity.com/paper/1908.01091/) (2019)
- [Sequential mastery of multiple visual tasks: Networks naturally learn to learn and forget to forget](https://mlsanity.com/paper/1905.10837/) (2019)
- [A Tale of Three Probabilistic Families: Discriminative, Descriptive and Generative Models](https://mlsanity.com/paper/1810.04261/) (2018)
- [BinPlay: A Binary Latent Autoencoder for Generative Replay Continual Learning](https://mlsanity.com/paper/2011.14960/) (2020)
- [Dissecting Catastrophic Forgetting in Continual Learning by Deep Visualization](https://mlsanity.com/paper/2001.01578/) (2020)

## Built on by (most cited first, 5 of 5)

- [Memorization Without Overfitting: Analyzing the Training Dynamics of Large Language Models](https://mlsanity.com/paper/2205.10770/) (2022)
- [Learning Data Representations with Joint Diffusion Models](https://mlsanity.com/paper/2301.13622/) (2023)
- [Joint Diffusion models in Continual Learning](https://mlsanity.com/paper/2411.08224/) (2024)
- [One-Shot Machine Unlearning with Mnemonic Code](https://mlsanity.com/paper/2306.05670/) (2023)
- [Optimisation Is Not What You Need](https://mlsanity.com/paper/2507.03045/) (2025)

## Abstract (arXiv)

Catastrophic forgetting of previously learned knowledge while learning new tasks is a widely observed limitation of contemporary neural networks. Although many continual learning methods are proposed to mitigate this drawback, the main question remains unanswered: what is the root cause of catastrophic forgetting? In this work, we aim at answering this question by posing and validating a set of research hypotheses related to the specificity of representations built internally by neural models. More specifically, we design a set of empirical evaluations that compare the robustness of representations in discriminative and generative models against catastrophic forgetting. We observe that representations learned by discriminative models are more prone to catastrophic forgetting than their generative counterparts, which sheds new light on the advantages of developing generative models for continual learning. Finally, our work opens new research pathways and possibilities to adopt generative models in continual learning beyond mere replay mechanisms.