Michael Hersche

My research focuses on designing intelligent systems for abstract reasoning in both the visual and natural language domains.

Short Bio:

Michael Hersche earned his B.S. in Electrical Engineering from the University of Applied Sciences Zurich in 2015 and an M.Sc. in Electrical Engineering and Information Technology from ETH Zürich in 2018. He completed his doctoral studies at ETH Zürich in 2023 under the supervision of Prof. Dr. Luca Benini. During his doctoral studies, he worked at the Integrated Systems Laboratory, ETH Zürich, before joining the In-Memory Computing Group at IBM Research-Zurich in 2021, where he now serves as a Research Associate. Dr. Hersche is a recipient of the 2020 IBM Ph.D. Fellowship Award and received the ETH Medal for his distinguished doctoral thesis.

Selected Publications:

• A. Terzić, M. Hersche, G. Camposampiero, T. Hofmann, A. Sebastian, A. Rahimi, 'On the Expressiveness and Length Generalization of Selective State Space Models on Regular Languages', 39th Annual AAAI Conference on Artificial Intelligence, 2025.
• J. Thomm, G. Camposampiero, A. Terzic, M. Hersche, B. Schölkopf, A. Rahimi, 'Limits of transformer language models on learning to compose algorithms', Conference on Neural Information Processing Systems (NeurIPS), 2024.
• M. Hersche, A. Terzić, G. Karunaratne, J. Langenegger, A. Pouget, G. Cherubini, L. Benini, A. Sebastian, and A. Rahimi, 'Factorizers for Distributed Sparse Block Codes', Neurosymbolic Artificial Intelligence, 2024.
• N. Menet, M. Hersche, K. Karunaratne, L. Benini, A. Sebastian, A. Rahimi, 'MIMONets: multiple-input-multiple-output neural networks exploiting computation in superposition', Conference on Neural Information Processing Systems (NeurIPS), 2023.
• M. Hersche, M. Zeqiri, L. Benini, A. Sebastian, A. Rahimi, 'A neuro-vector-symbolic architecture for solving Raven’s progressive matrices', Nature Machine Intelligence, 2023.
• J. Langenegger, G. Karunaratne, M. Hersche, L. Benini, A. Sebastian, A. Rahimi, 'In-memory factorization of holographic perceptual representations', Nature Nanotechnology, 2023.
• M. Hersche, G. Karunaratne, G. Cherubini, L. Benini, A. Sebastian, A. Rahimi, 'Constrained few-shot class-incremental learning', Conference on Computer Vision and Pattern Recognition (CVPR), 2022.
• M. Hersche, S. Lippuner, M. Korb, L. Benini, and A. Rahimi, 'Near-channel Classifier: Symbiotic Communication and Classification in High-dimensional Space', Brain Informatics, 2021.
• T. Ingolfsson, M. Hersche, X. Wang, N. Kobayashi, L. Cavigelli, and L. Benini, 'EEG-TCNet: An Accurate Temporal Convolutional Network for Embedded Motor-Imagery Brain–Machine Interfaces', IEEE International Conference on Systems, Man, and Cybernetics, 2020.