About me
I am a PhD student at Bar Ilan University, working under the joint supervision of Prof. Gal Chechik, head of the Computational Neurobiology Lab at the Gonda Multidisciplinary Brain Research Center, and Prof. Ethan Fetaya from the Faculty of Engineering. My main research interests are multitask learning and equivariant weight space networks.
News
- Multi-head Efficient Decoding for Transformer-based ASR accepted at ICASSP 2025.
- Keyword-Guided Adaptation of Automatic Speech Recognition accepted at Interspeech 2024.
- Two papers to be presented at ICML 2024:
- Open-vocabulary Keyword-spotting with Adaptive Instance Normalization accepted at ICASSP 2024.
- Two papers accepted at ICML 2023:
- Multi-Task Learning as a Bargaining Game has been accepted at ICML 2022.
- Personalized Federated Learning with Gaussian Processes accepted to NeurIPS 2021.
- Two papers to be presented at ICML 2021:
- Two papers to be presented at ICLR 2021:
Publications
Whisper in Medusa’s Ear: Multi-head Efficient Decoding for Transformer-based ASR
Y. Segal-Feldman, A. Shamsian, A. Navon, G. Hetz, J. Keshet
International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025).
Keyword-Guided Adaptation of Automatic Speech Recognition
A. Shamsian, A. Navon, N. Glazer, G. Hetz, J. Keshet
Conference of the International Speech Communication Association (InterSpeech 2024).
Equivariant Deep Weight Space Alignment
A. Navon, A. Shamsian, E. Fetaya, G. Chechik, N. Dym, H. Maron
International Conference on Machine Learning, ICML, 2024.
Improved Generalization of Weight Space Networks via Augmentations
A. Shamsian, A. Navon, D. Zhang, Y. Zhang, E. Fetaya, G. Chechik, H. Maron
International Conference on Machine Learning, ICML, 2024.
Open-vocabulary Keyword-spotting with Adaptive Instance Normalization
A. Navon, A. Shamsian, N. Glazer, G. Hetz, J. Keshet
International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2024).
Equivariant Architectures for Learning in Deep Weight Spaces
A. Navon, A. Shamsian, I. Achituve, E. Fetaya, G. Chechik, H. Maron
International Conference on Machine Learning, ICML, 2023.
Oral presentation.
Auxiliary Learning as an Asymmetric Bargaining Game
A. Shamsian, A. Navon, N. Glazer, K. Kawaguchi, G. Chechik, E. Fetaya
International Conference on Machine Learning, ICML, 2023.
Multi-Task Learning as a Bargaining Game
A. Navon, A. Shamsian, I. Achituve, H. Maron, K. Kawaguchi, G. Chechik, E. Fetaya
International Conference on Machine Learning, ICML, 2022.
Personalized Federated Learning using Hypernetworks
A. Shamsian, A. Navon, G. Chechik, E. Fetaya
International Conference on Machine Learning, ICML, 2021.
Personalized Federated Learning with Gaussian Processes
I. Achituve, A. Shamsian, A. Navon, G. Chechik, E. Fetaya
Neural Information Processing Systems, NeurIPS 2021.
GP-Tree: A Gaussian Process Classifier for Few-Shot Incremental Learning
I. Achituve, A. Navon, Y. Yemini, G. Chechik, E. Fetaya
International Conference on Machine Learning, ICML 2021.
Learning the Pareto Front with Hypernetworks
A. Navon, A. Shamsian, G. Chechik, E. Fetaya
International Conference on Learning Representations, ICLR 2021.
Auxiliary Learning by Implicit Differentiation
A. Navon, I. Achituve, H. Maron, G. Chechik, E. Fetaya
International Conference on Learning Representations, ICLR 2021.
Capturing Between-Tasks Covariance and Similarities Using Multivariate Linear Mixed Models
A. Navon, S. Rosset
Electronic Journal of Statistics, EJS 2020.
Preprints
Learning Discrete Weights and Activations Using the Local Reparameterization Trick
G. Berger, A. Navon, E. Fetaya
Preprint, 2023.
A Study on the Evaluation of Generative Models
E. Betzalel, C. Penso, A. Navon, E. Fetaya
Preprint, 2022.