# Understanding Continual Learning Settings with Data Distribution Drift Analysis

arXiv: https://arxiv.org/abs/2104.01678 · HTML: https://mlsanity.com/paper/2104.01678/

## Key facts

- Cited by 38 papers in the map (data through 14 Oct 2026).
- Builds on 44 of its 76 references that are in the map.
- First posted on arXiv 4 Apr 2021.

## Builds on (most cited first, 30 of 44)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Invariant Risk Minimization](https://mlsanity.com/paper/1907.02893/) (2019)
- [Progressive Neural Networks](https://mlsanity.com/paper/1606.04671/) (2016)
- [Continual Learning with Deep Generative Replay](https://mlsanity.com/paper/1705.08690/) (2017)
- [Efficient Lifelong Learning with A-GEM](https://mlsanity.com/paper/1812.00420/) (2018)
- [Meta-Learning in Neural Networks: A Survey](https://mlsanity.com/paper/2004.05439/) (2020)
- [Out-of-Distribution Generalization via Risk Extrapolation (REx)](https://mlsanity.com/paper/2003.00688/) (2020)
- [Large Scale Incremental Learning](https://mlsanity.com/paper/1905.13260/) (2019)
- [Sim-to-Real Transfer of Robotic Control with Dynamics Randomization](https://mlsanity.com/paper/1710.06537/) (2017)
- [Deep Visual Domain Adaptation: A Survey](https://mlsanity.com/paper/1802.03601/) (2018)
- [Three scenarios for continual learning](https://mlsanity.com/paper/1904.07734/) (2019)
- [Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interference](https://mlsanity.com/paper/1810.11910/) (2018)
- [Progress & Compress: A scalable framework for continual learning](https://mlsanity.com/paper/1805.06370/) (2018)
- [Online Continual Learning with Maximally Interfered Retrieval](https://mlsanity.com/paper/1908.04742/) (2019)
- [PathNet: Evolution Channels Gradient Descent in Super Neural Networks](https://mlsanity.com/paper/1701.08734/) (2017)
- [CORe50: a New Dataset and Benchmark for Continuous Object Recognition](https://mlsanity.com/paper/1705.03550/) (2017)
- [Online Learning: A Comprehensive Survey](https://mlsanity.com/paper/1802.02871/) (2018)
- [Online Structured Laplace Approximations For Overcoming Catastrophic Forgetting](https://mlsanity.com/paper/1805.07810/) (2018)
- [Modeling the Background for Incremental Learning in Semantic Segmentation](https://mlsanity.com/paper/2002.00718/) (2020)
- [Meta-Learning Representations for Continual Learning](https://mlsanity.com/paper/1905.12588/) (2019)
- [Invariant Risk Minimization Games](https://mlsanity.com/paper/2002.04692/) (2020)
- [PLOP: Learning without Forgetting for Continual Semantic Segmentation](https://mlsanity.com/paper/2011.11390/) (2020)
- [Sim-to-Real via Sim-to-Sim: Data-efficient Robotic Grasping via Randomized-to-Canonical Adaptation Networks](https://mlsanity.com/paper/1812.07252/) (2018)
- [A Closer Look at Invalid Action Masking in Policy Gradient Algorithms](https://mlsanity.com/paper/2006.14171/) (2020)
- [Continual Prototype Evolution: Learning Online from Non-Stationary Data Streams](https://mlsanity.com/paper/2009.00919/) (2020)
- [iTAML: An Incremental Task-Agnostic Meta-learning Approach](https://mlsanity.com/paper/2003.11652/) (2020)
- [Optimal Continual Learning has Perfect Memory and is NP-hard](https://mlsanity.com/paper/2006.05188/) (2020)

## Built on by (most cited first, 30 of 38)

- [New Insights on Reducing Abrupt Representation Change in Online Continual Learning](https://mlsanity.com/paper/2104.05025/) (2021)
- [Continual Pre-Training of Large Language Models: How to (re)warm your model?](https://mlsanity.com/paper/2308.04014/) (2023)
- [Simple and Scalable Strategies to Continually Pre-train Large Language Models](https://mlsanity.com/paper/2403.08763/) (2024)
- [Reuse, Don't Retrain: A Recipe for Continued Pretraining of Language Models](https://mlsanity.com/paper/2407.07263/) (2024)
- [Continual Learning via Local Module Composition](https://mlsanity.com/paper/2111.07736/) (2021)
- [Continual Learning with Foundation Models: An Empirical Study of Latent Replay](https://mlsanity.com/paper/2205.00329/) (2022)
- [Unveiling the Secret Recipe: A Guide For Supervised Fine-Tuning Small LLMs](https://mlsanity.com/paper/2412.13337/) (2024)
- [CLAD: A realistic Continual Learning benchmark for Autonomous Driving](https://mlsanity.com/paper/2210.03482/) (2022)
- [Graph Lifelong Learning: A Survey](https://mlsanity.com/paper/2202.10688/) (2022)
- [CLEVA-Compass: A Continual Learning EValuation Assessment Compass to Promote Research Transparency and Comparability](https://mlsanity.com/paper/2110.03331/) (2021)
- [Online-LoRA: Task-free Online Continual Learning via Low Rank Adaptation](https://mlsanity.com/paper/2411.05663/) (2024)
- [Challenging Common Assumptions about Catastrophic Forgetting](https://mlsanity.com/paper/2207.04543/) (2022)
- [Continual Learning in Deep Networks: an Analysis of the Last Layer](https://mlsanity.com/paper/2106.01834/) (2021)
- [Tackling Catastrophic Forgetting and Background Shift in Continual Semantic Segmentation](https://mlsanity.com/paper/2106.15287/) (2021)
- [Towards Effective and Efficient Continual Pre-training of Large Language Models](https://mlsanity.com/paper/2407.18743/) (2024)
- [Handling Out-of-Distribution Data: A Survey](https://mlsanity.com/paper/2507.21160/) (2025)
- [Robust Fast Adaptation from Adversarially Explicit Task Distribution Generation](https://mlsanity.com/paper/2407.19523/) (2024)
- [Task-Agnostic Continual Reinforcement Learning: Gaining Insights and Overcoming Challenges](https://mlsanity.com/paper/2205.14495/) (2022)
- [Beyond Supervised Continual Learning: a Review](https://mlsanity.com/paper/2208.14307/) (2022)
- [Continual Learning for Predictive Maintenance: Overview and Challenges](https://mlsanity.com/paper/2301.12467/) (2023)
- [Preventing Catastrophic Forgetting in Continual Learning of New Natural Language Tasks](https://mlsanity.com/paper/2302.11074/) (2023)
- [A Simple Yet Effective Strategy to Robustify the Meta Learning Paradigm](https://mlsanity.com/paper/2310.00708/) (2023)
- [Continual Learning Under Language Shift](https://mlsanity.com/paper/2311.01200/) (2023)
- [Continual Learning for Unsupervised Anomaly Detection in Continuous Auditing of Financial Accounting Data](https://mlsanity.com/paper/2112.13215/) (2021)
- [Continual Feature Selection: Spurious Features in Continual Learning](https://mlsanity.com/paper/2203.01012/) (2022)
- [Federated Continual Learning to Detect Accounting Anomalies in Financial Auditing](https://mlsanity.com/paper/2210.15051/) (2022)
- [A Compositional Approach to Creating Architecture Frameworks with an Application to Distributed AI Systems](https://mlsanity.com/paper/2212.13570/) (2022)
- [Enhancing PPO with Trajectory-Aware Hybrid Policies](https://mlsanity.com/paper/2502.15968/) (2025)
- [A Study of Continual Learning Methods for Q-Learning](https://mlsanity.com/paper/2206.03934/) (2022)
- [Revisiting Weight Regularization for Low-Rank Continual Learning](https://mlsanity.com/paper/2602.17559/) (2026)

## Abstract (arXiv)

Classical machine learning algorithms often assume that the data are drawn i.i.d. from a stationary probability distribution. Recently, continual learning emerged as a rapidly growing area of machine learning where this assumption is relaxed, i.e. where the data distribution is non-stationary and changes over time. This paper represents the state of data distribution by a context variable c. A drift in c leads to a data distribution drift. A context drift may change the target distribution, the input distribution, or both. Moreover, distribution drifts might be abrupt or gradual. In continual learning, context drifts may interfere with the learning process and erase previously learned knowledge; thus, continual learning algorithms must include specialized mechanisms to deal with such drifts. In this paper, we aim to identify and categorize different types of context drifts and potential assumptions about them, to better characterize various continual-learning scenarios. Moreover, we propose to use the distribution drift framework to provide more precise definitions of several terms commonly used in the continual learning field.