Skip to main content

UNDERSTANDING THE ROBUSTNESS OF SELF-SUPERVISED LEARNING THROUGH TOPIC MODELING

Publication ,  Conference
Luo, Z; Wu, S; Weng, C; Zhou, M; Ge, R
Published in: 11th International Conference on Learning Representations, ICLR 2023
January 1, 2023

Self-supervised learning has significantly improved the performance of many NLP tasks. However, how can self-supervised learning discover useful representations, and why is it better than traditional approaches such as probabilistic models are still largely unknown. In this paper, we focus on the context of topic modeling and highlight a key advantage of self-supervised learning - when applied to data generated by topic models, self-supervised learning can be oblivious to the specific model, and hence is less susceptible to model misspecification. In particular, we prove that commonly used self-supervised objectives based on reconstruction or contrastive samples can both recover useful posterior information for general topic models. Empirically, we show that the same objectives can perform on par with posterior inference using the correct model, while outperforming posterior inference using misspecified models.

Duke Scholars

Published In

11th International Conference on Learning Representations, ICLR 2023

Publication Date

January 1, 2023
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Luo, Z., Wu, S., Weng, C., Zhou, M., & Ge, R. (2023). UNDERSTANDING THE ROBUSTNESS OF SELF-SUPERVISED LEARNING THROUGH TOPIC MODELING. In 11th International Conference on Learning Representations, ICLR 2023.
Luo, Z., S. Wu, C. Weng, M. Zhou, and R. Ge. “UNDERSTANDING THE ROBUSTNESS OF SELF-SUPERVISED LEARNING THROUGH TOPIC MODELING.” In 11th International Conference on Learning Representations, ICLR 2023, 2023.
Luo Z, Wu S, Weng C, Zhou M, Ge R. UNDERSTANDING THE ROBUSTNESS OF SELF-SUPERVISED LEARNING THROUGH TOPIC MODELING. In: 11th International Conference on Learning Representations, ICLR 2023. 2023.
Luo, Z., et al. “UNDERSTANDING THE ROBUSTNESS OF SELF-SUPERVISED LEARNING THROUGH TOPIC MODELING.” 11th International Conference on Learning Representations, ICLR 2023, 2023.
Luo Z, Wu S, Weng C, Zhou M, Ge R. UNDERSTANDING THE ROBUSTNESS OF SELF-SUPERVISED LEARNING THROUGH TOPIC MODELING. 11th International Conference on Learning Representations, ICLR 2023. 2023.

Published In

11th International Conference on Learning Representations, ICLR 2023

Publication Date

January 1, 2023