# Class Gradient Projection For Continual Learning

arXiv: https://arxiv.org/abs/2311.14905 · HTML: https://mlsanity.com/paper/2311.14905/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 48 references that are in the map.
- First posted on arXiv 25 Nov 2023.

## Builds on (most cited first, 30 of 34)

- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms](https://mlsanity.com/paper/1708.07747/) (2017)
- [Exploring Simple Siamese Representation Learning](https://mlsanity.com/paper/2011.10566/) (2020)
- [Supervised Contrastive Learning](https://mlsanity.com/paper/2004.11362/) (2020)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence](https://mlsanity.com/paper/2001.07685/) (2020)
- [Gradient Episodic Memory for Continual Learning](https://mlsanity.com/paper/1706.08840/) (2017)
- [Progressive Neural Networks](https://mlsanity.com/paper/1606.04671/) (2016)
- [Memory Aware Synapses: Learning what (not) to forget](https://mlsanity.com/paper/1711.09601/) (2017)
- [Efficient Lifelong Learning with A-GEM](https://mlsanity.com/paper/1812.00420/) (2018)
- [AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty](https://mlsanity.com/paper/1912.02781/) (2019)
- [Overcoming catastrophic forgetting with hard attention to the task](https://mlsanity.com/paper/1801.01423/) (2018)
- [Prototypical Contrastive Learning of Unsupervised Representations](https://mlsanity.com/paper/2005.04966/) (2020)
- [Orthogonal Gradient Descent for Continual Learning](https://mlsanity.com/paper/1910.07104/) (2019)
- [Gradient Projection Memory for Continual Learning](https://mlsanity.com/paper/2103.09762/) (2021)
- [Variational Continual Learning](https://mlsanity.com/paper/1710.10628/) (2017)
- [On Tiny Episodic Memories in Continual Learning](https://mlsanity.com/paper/1902.10486/) (2019)
- [Learn to Grow: A Continual Structure Learning Framework for Overcoming Catastrophic Forgetting](https://mlsanity.com/paper/1904.00310/) (2019)
- [Co2L: Contrastive Continual Learning](https://mlsanity.com/paper/2106.14413/) (2021)
- [Overcoming Catastrophic Forgetting by Incremental Moment Matching](https://mlsanity.com/paper/1703.08475/) (2017)
- [Compacting, Picking and Growing for Unforgetting Continual Learning](https://mlsanity.com/paper/1910.06562/) (2019)
- [Continual Learning of Context-dependent Processing in Neural Networks](https://mlsanity.com/paper/1810.01256/) (2018)
- [Adversarial Continual Learning](https://mlsanity.com/paper/2003.09553/) (2020)
- [Scalable and Order-robust Continual Learning with Additive Parameter Decomposition](https://mlsanity.com/paper/1902.09432/) (2019)
- [Conditional Channel Gated Networks for Task-Aware Continual Learning](https://mlsanity.com/paper/2004.00070/) (2020)
- [Efficient Continual Learning with Modular Networks and Task-Driven Priors](https://mlsanity.com/paper/2012.12631/) (2020)
- [TRGP: Trust Region Gradient Projection for Continual Learning](https://mlsanity.com/paper/2202.02931/) (2022)
- [Self-supervised Knowledge Distillation Using Singular Value Decomposition](https://mlsanity.com/paper/1807.06819/) (2018)

## Built on by (most cited first, 11 of 11)

- [CoIN: A Benchmark of Continual Instruction tuNing for Multimodel Large Language Model](https://mlsanity.com/paper/2403.08350/) (2024)
- [Deep Unlearning: Fast and Efficient Gradient-free Approach to Class Forgetting](https://mlsanity.com/paper/2312.00761/) (2023)
- [LW2G: Learning Whether to Grow for Prompt-based Continual Learning](https://mlsanity.com/paper/2409.18860/) (2024)
- [Towards Redundancy-Free Sub-networks in Continual Learning](https://mlsanity.com/paper/2312.00840/) (2023)
- [RegCL: Compact Continual SAM Adaptation for Visual Grounding in Multi-Sensorial Media](https://mlsanity.com/paper/2507.12297/) (2025)
- [Janus-LoRA: A Balanced Low-Rank Adaptation for Continual Learning](https://mlsanity.com/paper/2605.28495/) (2026)
- [Gradient Rewiring for Editable Graph Neural Network Training](https://mlsanity.com/paper/2410.15556/) (2024)
- [FALCON: Fine-grained Activation Manipulation by Contrastive Orthogonal Unalignment for Large Language Model](https://mlsanity.com/paper/2502.01472/) (2025)
- [Orthogonal Calibration for Asynchronous Federated Learning](https://mlsanity.com/paper/2502.15940/) (2025)
- [Hidden Failure Modes of Gradient Modification under Adam in Continual Learning, and Adaptive Decoupled Moment Routing as a Repair](https://mlsanity.com/paper/2604.22407/) (2026)
- [Recursive Binding on a Budget: Subspace Carving in Order-p Tensor Memories](https://mlsanity.com/paper/2606.11391/) (2026)

## Abstract (arXiv)

Catastrophic forgetting is one of the most critical challenges in Continual Learning (CL). Recent approaches tackle this problem by projecting the gradient update orthogonal to the gradient subspace of existing tasks. While the results are remarkable, those approaches ignore the fact that these calculated gradients are not guaranteed to be orthogonal to the gradient subspace of each class due to the class deviation in tasks, e.g., distinguishing "Man" from "Sea" v.s. differentiating "Boy" from "Girl". Therefore, this strategy may still cause catastrophic forgetting for some classes. In this paper, we propose Class Gradient Projection (CGP), which calculates the gradient subspace from individual classes rather than tasks. Gradient update orthogonal to the gradient subspace of existing classes can be effectively utilized to minimize interference from other classes. To improve the generalization and efficiency, we further design a Base Refining (BR) algorithm to combine similar classes and refine class bases dynamically. Moreover, we leverage a contrastive learning method to improve the model's ability to handle unseen tasks. Extensive experiments on benchmark datasets demonstrate the effectiveness of our proposed approach. It improves the previous methods by 2.0% on the CIFAR-100 dataset.