Interaction between Deep Learning and Neuroscience
Historically neuroscience has inspired many deep learning techniques. Now the two fields are converging. Neuroscience continues to inspire more powerful deep learning models and deep learning models are used to understand the working mechanism of the brain. Interaction between the two fields will benefit both of them.
Reinforcement Learning and Algorithmic Game Theory
Background: Reinforcement Learning enables the agent finishing several different tasks in unknown environments. Algorithmic Game Theory aims at finding good strategy for each agent in a multi-agent system under certain rules. Key Techniques: Conterfactual Regret Minimization, Thompson Sampling, Nash Equilibrium Finding, Model-based Reinforcement Learning, Combo Action.
Adversarial Attacks and Defenses (for Deep Learning)
Deep Neural networks (DNNs) are challenged by their vulnerability to adversarial examples, which are crafted by adding human-imperceptible noises to real examples, but make a model output inaccurate predictions. Researches on adversarial attacks and defenses are the foundations of building robust artificial intelligence systems.
Interpretable Machine Learning Techniques and Visualization