Cory earned his BS/MS degree in computer engineering from RIT in 2011 with a minor in mathematics. Currently, he is pursuing the PhD degree in microsystems engineering. His primary research interests include neuromorphic/neuromemristive circuit design and unconventional computing paradigms.
James’ research interests include data science and machine learning with an emphasis on computer vision and spatiotemporal data. He is currently designing a classification framework for still images. The framework utilizes a highly customized implementation of hierarchical temporal memory at its core. Future work includes expanding the framework to perform prediction and anomaly detection, especially with video streams.
Lennard's general research area is biologically-inspired computer architecture. His research interests include memory-based computing, and spatiotemporal data analysis. Lately, he has been focused on flash technology, bimodal (audio-visual) data, and hierarchical temporal memory. He is soon to join IBM in Austin.
Levs is accelerating HTM on a GPU, to deploy on personal computers and possibly portable devices. HTM is highly parallel, which allows to speed it up from ~100ms to ~100us per iteration. This makes many new applications computationally possible. Fast training also helps rapidly explore new applications. A GPU implementation can be a high-level model for a hardware implementation, where an iteration can take as little as 1us, while being drastically more efficient.
Qutaiba is interested in Reservoir computing and spatiotemporal signal processing. Currently, he is designing a scalable, reconfigurable and power efficient hardware architecture for Echo State Networks, a type of Reservoir Computing, using Memristive devices. Such architecture opens wide doors toward using Reservoir Computing in processing information at real-time speed in small and limited power resource applications such as therapeutic devices and body sensors.
Abdullah is working on developing a scalable, intelligent and self-learning hardware substrate inspired by the operating principles of the neocortex. The architecture models will be realized on FPGA Virtex 5 board and will target applications such as anomaly detection and natural language processing.
Abhishek's interest is in neuromorphic digital systems. He has designed an optimal digital hardware for Spike Time Dependent Plasticity Rule.
Daniel is pursuing his Master's degree in Electrical Engineering. His current research works are based on reservoir computing, which is inspired from cortical microcircuits of the brain. He is largely interested in designing power optimized RTL architectures for neural circuits.
Dan is researching the advantages of incorporating thalamic inspired computer architectures into existing neural based systems. Seldom has the community worked to incorporate subcortical theory of the lower brain to existing models, although evidence suggest an intimate relationship between process loops occurring in the neocortex and the subcortical structures to which they are paired. Thalamic inspired architectures
Amanda's general interests center around neuroscience and understanding the mechanisms underlying human cognition, as well as other emergent traits such as personality, memory, and intelligence. At NanoLab, Amanda is investigating potential applications for HTM. She is also looking into extending an existing mathematical framework for HTM, established by NanoLab alumnus James Mnatzaganian.