On the role of noise in factorizers for disentangling distributed representations
- 2024
- NeurIPS 2024
Geethan Karunaratne received his B.Sc. degree in Electronic and Telecommunication Engineering from the University of Moratuwa, Sri Lanka in 2014. He received the M.Sc./Ph.D. degrees in Information Technology and Electrical Engineering from ETH Zurich in 2018 and 2023 respectively. Before beginning his master's, he joined the then incubated start-up Paraqum Technologies, Sri Lanka, where he worked on developing high-end hardware video encoders and decoders.
In 2018, he joined IBM Research – Zurich, where he is currently a member of the In-memory computing group. Geethan is currently researching on designing next-generation in-memory computing hardware and implementing emerging computing paradigms on in-memory computing fabrics. His main research interests are in-memory computing and brain-inspired computing.
Geethan is interested in hardware acceleration, energy-efficient VLSI architectures, machine learning, neuromorphic computing, and machine vision. He has experience in configurable architectures, FPGA ASIC design flows, and hardware emulation.