Abstract
Catastrophic forgetting is a tough issue when the agent faces the sequential multi-task learning scenario without storing previous task information. It gradually becomes an obstacle to achieve artificial general intelligence which is generally believed to behave like a human with continuous learning capability. In this paper, we propose to utilize the variational Bayesian inference method to overcome catastrophic forgetting. By pruning the neural network according to the mean and variance of weights, parameters are vastly reduced, which mitigates the storage problem of double parameters required in variational Bayesian inference. Based on this lightweight version, autoencoders trained on different tasks are employed to self-adaptively match the corresponding task parameters to tackle sequential multi-task learning problem. We show experimentally on several fundamental datasets that the proposed method can perform substantial improvements without catastrophic forgetting over other classic methods especially in the setting where the probability distributions between tasks present more different.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Thrun, S.: Lifelong learning algorithms. In: Thrun, S., Pratt, L. (eds.) Learning to Learn, pp. 181–209. Springer, Boston (1998). https://doi.org/10.1007/978-1-4615-5529-2_8
Kirkpatrick, J., et al.: Overcoming catastrophic forgetting in neural networks. Proc. Natl. Acad. Sci. 114(13), 3521–3526 (2017)
Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 580–587 (2014)
Donahue, J., et al.: DeCAF: a deep convolutional activation feature for generic visual recognition. In: Proceedings of the International Conference on Machine Learning, pp. 647–655 (2014)
Li, Z., Hoiem, D.: Learning without forgetting. IEEE Trans. Pattern Anal. Mach. Intell. (2017)
Xiong, F., et al.: Guided policy search for sequential multitask learning. IEEE Trans. Syst. Man Cybern. Syst. (2018)
Fernando, C., et al.: PathNet: Evolution channels gradient descent in super neural networks. arXiv preprint arXiv:1701.08734 (2017)
Serrà , J., SurÃs, D., Miron, M., Karatzoglou, A.: Overcoming catastrophic forgetting with hard attention to the task. In: Proceedings of the International Conference on Machine Learning (2018)
Ep Triki, A.R., Aljundi, R., Blaschko, M., Tuytelaars, T.: Encoder based lifelong learning. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1320–1328 (2017)
Lee, S.W., Kim, J.H., Jun, J., Ha, J.W., Zhang, B.T.: Overcoming catastrophic forgetting by incremental moment matching. In: Advances in Neural Information Processing Systems, pp. 4655–4665 (2017)
Blundell, C., Cornebise, J., Kavukcuoglu, K., Wierstra, D.: Weight uncertainty in neural networks. Proc. Int. Conf. Mach. Learn. 37, 1613–1622 (2015)
Acknowledgments
This work was supported in part by the National Key Research and Development Plan of China under Grants 2017YFB1300202 and 2016YFC0300801, in part by the NSFC under Grants U1613213, 61627808, 61503383, 61210009, 91648205, 61702516, and 61473236, in part by the MOST under Grant 2015BAK35B00 and Grant 2015BAK35B01, in part by the Guangdong Science and Technology Department under Grant 2016B090910001.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Xiong, F., Liu, Z., Yang, X. (2018). Overcoming Catastrophic Forgetting with Self-adaptive Identifiers. In: Cheng, L., Leung, A., Ozawa, S. (eds) Neural Information Processing. ICONIP 2018. Lecture Notes in Computer Science(), vol 11303. Springer, Cham. https://doi.org/10.1007/978-3-030-04182-3_43
Download citation
DOI: https://doi.org/10.1007/978-3-030-04182-3_43
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-04181-6
Online ISBN: 978-3-030-04182-3
eBook Packages: Computer ScienceComputer Science (R0)