`

Timezone: »

 
Semi-supervised Multiple Instance Learning using Variational Auto-Encoders
Ali Nihat Uzunalioglu · Tameem Adel · Jakub M. Tomczak
Event URL: https://openreview.net/forum?id=irZt4xL3XeM »

We consider the multiple-instance learning (MIL) paradigm, which is a special case of supervised learning where training instances are grouped into bags. In MIL, the hidden instance labels do not have to be the same as the label of the comprising bag. On the other hand, the hybrid modelling approach is known to possess advantages basically due to the smooth consolidation of both discriminative and generative components. In this paper, we investigate whether we can get the best of both worlds (MIL and hybrid modelling), especially in a semi-supervised learning (SSL) setting. We first integrate a variational autoencoder (VAE), which is a powerful deep generative model, with an attention-based MIL classifier, then evaluate the performance of the resulting model in SSL. We assess the proposed approach on an established benchmark as well as a real-world medical dataset.

Author Information

Ali Nihat Uzunalioglu (Vrije Universiteit)
Tameem Adel (University of Cambridge)
Jakub M. Tomczak (Vrije Universiteit Amsterdam)

Jakub Tomczak is an assistant professor of Artificial Intelligence in the Computational Intelligence group at the Vrije Universiteit Amsterdam. From October 2016 to September 2018 he was a Marie Sklodowska-Curie Individual Fellow in Prof. Max Welling’s group at the University of Amsterdam. Afterwards, from October 2018 to October 2019, he worked as a deep learning researcher in Qualcomm AI Research Amsterdam. He obtained his Ph.D. in machine learning from the Wroclaw University of Technology (Poland). His research interests include probabilistic modeling, deep learning, approximate Bayesian modeling, and deep generative modeling (with special focus on Variational Auto-Encoders and flow-based models).

Related Events (a corresponding poster, oral, or spotlight)