Fifth International Workshop on Theoretical and Experimental Material Computing (TEMC 2024)
The Fifth International Workshop on Theoretical and Experimental Material Computing (TEMC 2024) will be held at Pohang University of Science and Technology (POSTECH), Pohang, Republic of Korea, as a satellite workshop of the Unconventional Computation and Natural Computation (UCNC 2024), 17–21 June 2024.
Material computing exploits unconventional physical substrates and/or unconventional computational models to perform physical computation in a non-silicon and/or non-Turing paradigm. Such computations find a natural home in a variety of unconventional computing applications, including sensing and real time systems, and unconventional computing materials, including magnetic materials. TEMC 2024 will encompass a range of theoretical and experimental approaches to material computing. The aim of the workshop is to bring together researchers from a range of connected fields, to inform of latest findings, to engage across the disciplines, to transfer discoveries and concepts from one field to another, and to inspire new collaborations and new ideas.
Call for Abstracts
Invited Talk: Quantum Neural Networks: Harnessing Natural Quantum Dynamics for Machine Learning
Speaker: Quoc Hoan Tran
Abstract: Quantum computing stands as a pivotal force in the exploration of unconventional computing territories. Reflecting on Feynman's perspective: given the significant difficulty of simulating quantum mechanics on classical computers, shouldn't we instead embrace the complexity and develop computers directly from quantum systems? Meanwhile, neural networks in classical systems play a central role in performing machine learning tasks. In this talk, inspired by both domains, I present the general structure of quantum neural networks that utilize natural quantum dynamics for machine learning tasks, particularly with high-dimensional and complex data that classical machine learning methods cannot handle well. I will discuss one of the major challenges in quantum computing: the inevitable buildup of quantum error in constructing quantum algorithms. Interestingly, quantum neural networks can help mitigate these quantum errors by training on noisy quantum data. Finally, I offer a complementary perspective to the current trends and approaches in quantum machine learning on leveraging prior knowledge about the data and quantum hardware to develop neural networks that exploit the inherent complexity of noisy intermediate-scale quantum devices.
Bio: Quoc Hoan Tran is a senior researcher at the Quantum Laboratory of Fujitsu Research, Fujitsu Limited. His current research focuses on novel architectures and intersecting applications of emerging quantum systems for machine learning within the unconventional computing framework. He received his predoctoral education and PhD from the University of Tokyo in 2020, with a background in applied mathematics and machine learning methods for dynamic systems. Before his current position at Fujitsu Research, he was a Postdoctoral Scholar and Specially Appointed Assistant Professor at the University of Tokyo in Professor Kohei Nakajima's group, working in the field of physical reservoir computing. He has also been a member of the MEXT - Quantum Leap Flagship Program (MEXT Q-LEAP) for the development of quantum software through intelligent quantum system design and its applications.
Programme and Organising Committee
Susan Stepney
David Griffin, Tian Gan, Alex McDonnell, Charles Swindells, Ian Vidamour
Simon O’Keefe, Martin Trefzer
Matt Ellis, Tom Hayward, Luca Manneschi, Eleni Vasilaki
TEMC 2024 is sponsored by the EPRSC MARCH project EP/V006029/1, EP/V006339/1, the University of York, UK, and the University of Sheffield, UK.
