Calibration of Continual Learning Models
Abstract: Continual Learning (CL) focuses on maximizing the predictive performance of a model across a non-stationary stream of data. Unfortunately, CL models tend to forget previous knowledge, thus often underperforming when compared with an offline model trained jointly on the entire data stream. Given that any CL model will eventually make mistakes, it is of crucial importance to build calibrated CL models: models that can reliably tell their confidence when making a prediction. Model calibration is an active research topic in machine learning, yet to be properly investigated in CL. We provide the first empirical study of the behavior of calibration approaches in CL, showing that CL strategies do not inherently learn calibrated models. To mitigate this issue, we design a continual calibration approach that improves the performance of post-processing calibration methods over a wide range of different benchmarks and CL strategies. CL does not necessarily need perfect predictive models, but rather it can benefit from reliable predictive models. We believe our study on continual calibration represents a first step towards this direction.
- An optimistic perspective on offline reinforcement learning. In Proceedings of the 37th International Conference on Machine Learning, ICML 2020, 13-18 July 2020, Virtual Event, pages 104–114. PMLR, 2020.
- The arcade learning environment: An evaluation platform for general agents. Journal of Artificial Intelligence Research, 47:253–279, 2013.
- Class-Incremental Continual Learning into the eXtended DER-verse. IEEE Transactions on Pattern Analysis and Machine Intelligence, pages 1–16, 2022.
- Dark Experience for General Continual Learning: A Strong, Simple Baseline. In Advances in Neural Information Processing Systems, pages 15920–15930. Curran Associates, Inc., 2020.
- Avalanche: A PyTorch Library for Deep Continual Learning. Journal of Machine Learning Research, 24(363):1–6, 2023.
- Learning with Rejection. In Algorithmic Learning Theory, pages 67–82, Cham, 2016. Springer International Publishing.
- A continual learning survey: Defying forgetting in classification tasks. IEEE Transactions on Pattern Analysis and Machine Intelligence, pages 1–1, 2021.
- Learning in Nonstationary Environments: A Survey. IEEE Computational Intelligence Magazine, 10(4):12–25, 2015.
- Robert French. Catastrophic forgetting in connectionist networks. Trends in Cognitive Sciences, 3(4):128–135, 1999.
- On calibration of modern neural networks. In Proceedings of the 34th International Conference on Machine Learning - Volume 70, pages 1321–1330, Sydney, NSW, Australia, 2017. JMLR.org.
- Replay in Deep Learning: Current Approaches and Missing Biological Elements. Neural computation, 33(11):2908–2950, 2021.
- Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016.
- Introducing eurosat: A novel dataset and deep learning benchmark for land use and land cover classification. In IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium, pages 204–207. IEEE, 2018.
- Eurosat: A novel dataset and deep learning benchmark for land use and land cover classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019.
- Continual learning for robotics: Definition, framework, learning strategies, opportunities and challenges. Information Fusion, 58:52–68, 2020.
- Avalanche: An End-to-End Library for Continual Learning. In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 3595–3605. IEEE, 2021.
- Gradient Episodic Memory for Continual Learning. In Advances in Neural Information Processing Systems. Curran Associates, Inc., 2017.
- Revisiting the arcade learning environment: Evaluation protocols and open problems for general agents. Journal of Artificial Intelligence Research, 61:523–562, 2018.
- Calibrated Model-Based Deep Reinforcement Learning. In Proceedings of the 36th International Conference on Machine Learning, pages 4314–4323. PMLR, 2019.
- Human-level control through deep reinforcement learning. Nat., 518(7540):529–533, 2015.
- Regularizing Neural Networks by Penalizing Confident Output Distributions. In ICLR Workshop, 2017.
- iCaRL: Incremental Classifier and Representation Learning. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 2001–2010, 2017.
- Experience Replay for Continual Learning. In Advances in Neural Information Processing Systems. Curran Associates, Inc., 2019.
- Evaluating model calibration in classification. In Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics, pages 3459–3467. PMLR, 2019.
- Gido M. van de Ven and Andreas S. Tolias. Three scenarios for continual learning. Continual Learning Workshop NeurIPS, 2018.
- A Survey on Learning to Reject. Proceedings of the IEEE, 111(2):185–215, 2023.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.