Self-supervised learning for neural topic models with variance–invariance–covariance regularization
Open Access
- 24 February 2025
- journal article
- research article
- Published by Springer Nature in Knowledge and Information Systems
- Vol. 67 (6) , 5057-5075
Abstract
In our study, we propose a self-supervised neural topic model (NTM) that combines the power of NTMs and regularized self-supervised learning methods to improve performance. NTMs use neural networks to learn latent topics hidden behind the words in documents, enabling greater flexibility and the ability to estimate more coherent topics compared to traditional topic models. On the other hand, some self-supervised learning methods use a joint embedding architecture with two identical networks that produce similar representations for two augmented versions of the same input. Regularizations are applied to these representations to prevent collapse, which would otherwise result in the networks outputting constant or redundant representations for all inputs. Our model enhances topic quality by explicitly regularizing latent topic representations of anchor and positive samples. We also introduced an adversarial data augmentation method to replace the heuristic sampling method. We further developed several variation models including those on the basis of an NTM that incorporates contrastive learning with both positive and negative samples. Experimental results on three datasets showed that our models outperformed baselines and state-of-the-art models both quantitatively and qualitatively.Keywords
All Related Versions
Funding Information
- Hiroshima University
This publication has 19 references indexed in Scilit:
- A survey on neural topic models: methods, applications, and challengesArtificial Intelligence Review, 2024
- TeachAugment: Data Augmentation Optimization Using Teacher Knowledge2022
- Mitigating Data Sparsity for Short Text Topic Modeling by Topic-Semantic Contrastive Learning2022
- Word Embedding-Based Topic Similarity Measures2021
- Pre-training is a Hot Topic: Contextualized Document Embeddings Improve Topic Coherence2021
- Improving Neural Topic Models using Knowledge Distillation2020
- Optuna2019
- Neural Relational Topic Models for Scientific Article Analysis2018
- Probabilistic topic modelsCommunications of the ACM, 2012
- NewsWeeder: Learning to Filter Netnews1995