Self-Supervised Learning for Robust Multimodal Neural  Networks

Mr.E.Sivarajan,Mr.V.Mouliraj,Mrs.M.Ramya,Mr.Saravanakumar Pichumani

Self-Supervised Learning for Robust Multimodal Neural Networks

Authors

Mr.E.Sivarajan,Mr.V.Mouliraj,Mrs.M.Ramya,Mr.Saravanakumar Pichumani

Keywords:

Self-supervision, representations for downstream tasks, requiring labels, for improving robustness and uncertainty estimation

Abstract

Self-supervision provides effective representations for downstream tasks withoutrequiring labels. However, existing approaches lag behind fully supervised training and are oftennot thought beneficial beyond obviating or reducing the need for annotations. We find that selfsupervision can benefit robustness in a variety of ways

References

. Anish Athalye, Nicholas Carlini, and David Wagner. Obfuscated gradients give a false sense of security: Circumventing defenses to adversarial examples. In Proceedings of the 35th International Conference on Machine Learning, ICML 2023, July 2023.

. Jens Behrmann, Will Grathwohl, Ricky T. Q. Chen, David Duvenaud, and Jörn-Henrik Jacobsen. Invertible residual networks. ArXiv, abs/1811.00995, 2024.

Downloads

Published

2024-08-21

How to Cite

Mr.E.Sivarajan,Mr.V.Mouliraj,Mrs.M.Ramya,Mr.Saravanakumar Pichumani. (2024). Self-Supervised Learning for Robust Multimodal Neural Networks. Journal of Computational Analysis and Applications (JoCAAA), 33(08), 3323–3329. Retrieved from https://eudoxuspress.com/index.php/pub/article/view/2476