R&D with a focus on development of algorithms and pipelines for real-time feature extraction and manipulation of spatio-temporal data to improve perception.

I worked as a postdoctoral fellow on the Green Brain Project with Prof. Thomas Nowotny, in collaboration with University of Sheffield. My role in the project was to build computational models of honeybee olfactory pathways that are related to reinforcement learning and decision making, and to enable this model to interact with the real world using chemosensors. The goal of the Green Brain Project was to build an autonomous flying robot with the navigational and learning abilities of a honeybee.I have also been a developer of the GeNN simulation environment which facilitates building spiking neuron networks on GPUs for accelerating their simulations (up to 200x speed-up compared to a CPU).

I studied visual cortical dynamics revealed by voltage sensitive dye imaging in-vivo, under the supervision of Yves Frégnac and Cyril Monier at UNIC, CNRS Gif-sur-Yvette. I performed experiments and developed a novel hybrid data analysis method to denoise the data and to identify stimulus-specific spatiotemporal activity patterns, using a combination of method-based GLM and data-driven PCA/ICA methods. This method provided better denoising than conventional methods, and revealed that neuronal populations exhibit faster and more precise responses compared to single neurons. My research was funded by Ecole des Neurosciences de Paris Graduate Program, and European BrainScaleS project. I defended my thesis in October 2012 and obtained my PhD degree in neuroscience.