# Hyperparameter Selection in Continual Learning

arXiv: https://arxiv.org/abs/2404.06466 · HTML: https://mlsanity.com/paper/2404.06466/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 34 references that are in the map.
- First posted on arXiv 9 Apr 2024.

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

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Continual Lifelong Learning with Neural Networks: A Review](https://mlsanity.com/paper/1802.07569/) (2018)
- [Efficient Lifelong Learning with A-GEM](https://mlsanity.com/paper/1812.00420/) (2018)
- [Dark Experience for General Continual Learning: a Strong, Simple Baseline](https://mlsanity.com/paper/2004.07211/) (2020)
- [A Comprehensive Survey of Continual Learning: Theory, Method and Application](https://mlsanity.com/paper/2302.00487/) (2023)
- [Three scenarios for continual learning](https://mlsanity.com/paper/1904.07734/) (2019)
- [Gradient based sample selection for online continual learning](https://mlsanity.com/paper/1903.08671/) (2019)
- [Online Continual Learning with Maximally Interfered Retrieval](https://mlsanity.com/paper/1908.04742/) (2019)
- [CORe50: a New Dataset and Benchmark for Continuous Object Recognition](https://mlsanity.com/paper/1705.03550/) (2017)
- [Re-evaluating Continual Learning Scenarios: A Categorization and Case for Strong Baselines](https://mlsanity.com/paper/1810.12488/) (2018)
- [New Insights on Reducing Abrupt Representation Change in Online Continual Learning](https://mlsanity.com/paper/2104.05025/) (2021)
- [Understanding the Role of Training Regimes in Continual Learning](https://mlsanity.com/paper/2006.06958/) (2020)
- [Class-Incremental Continual Learning into the eXtended DER-verse](https://mlsanity.com/paper/2201.00766/) (2022)
- [Continual Learning in Low-rank Orthogonal Subspaces](https://mlsanity.com/paper/2010.11635/) (2020)
- [Online Continual Learning with Natural Distribution Shifts: An Empirical Study with Visual Data](https://mlsanity.com/paper/2108.09020/) (2021)
- [Online Hyperparameter Optimization for Class-Incremental Learning](https://mlsanity.com/paper/2301.05032/) (2023)
- [Error Sensitivity Modulation based Experience Replay: Mitigating Abrupt Representation Drift in Continual Learning](https://mlsanity.com/paper/2302.11344/) (2023)
- [Defining Benchmarks for Continual Few-Shot Learning](https://mlsanity.com/paper/2004.11967/) (2020)
- [Approximate Bayesian Class-Conditional Models under Continuous Representation Shift](https://mlsanity.com/paper/2305.19076/) (2023)
- [Renate: A Library for Real-World Continual Learning](https://mlsanity.com/paper/2304.12067/) (2023)
- [Adaptive Hyperparameter Optimization for Continual Learning Scenarios](https://mlsanity.com/paper/2403.07015/) (2024)
- [Obeying the Order: Introducing Ordered Transfer Hyperparameter Optimisation](https://mlsanity.com/paper/2306.16916/) (2023)
- [What Can AutoML Do For Continual Learning?](https://mlsanity.com/paper/2311.11963/) (2023)

## Built on by (most cited first, 2 of 2)

- [No Forgetting Learning: Buffer-free Continual Learning Classification](https://mlsanity.com/paper/2503.04638/) (2025)
- [Shapley Neuron Values for Continual Learning: Which Neurons Matter Most?](https://mlsanity.com/paper/2605.15877/) (2026)

## Abstract (arXiv)

In continual learning (CL) -- where a learner trains on a stream of data -- standard hyperparameter optimisation (HPO) cannot be applied, as a learner does not have access to all of the data at the same time. This has prompted the development of CL-specific HPO frameworks. The most popular way to tune hyperparameters in CL is to repeatedly train over the whole data stream with different hyperparameter settings. However, this end-of-training HPO is unusable in practice since a learner can only see the stream once. Hence, there is an open question: what HPO framework should a practitioner use for a CL problem in reality? This paper looks at this question by comparing several realistic HPO frameworks. We find that none of the HPO frameworks considered, including end-of-training HPO, perform consistently better than the rest on popular CL benchmarks. We therefore arrive at a twofold conclusion: a) to be able to discriminate between HPO frameworks there is a need to move beyond the current most commonly used CL benchmarks, and b) on the popular CL benchmarks examined, a CL practitioner should use a realistic HPO framework and can select it based on factors separate from performance, for example compute efficiency.