Projects related to continual/lifelong learning in deep neural networks

Deep neural networks have demonstrated remarkable success in tasks that range from computer vision and natural language processing to robot control. Yet, in contrast to humans, when presented sequentially with new tasks they often forget completely how to solve their previously learned tasks. During this internship we will be looking into various research and innovation projects related to mitigating this problem. 

Good programming (C++ or Python) and deep learning (e.g., PyTorch) skills. Knowledge of deep reinforcement learning, neuroscience, or mobile app development will be considered an advantage.