Machine Learning and

Neural Computation

We pursue a broad range of topics in machine learning, from statistical theory and methodology to applications.

Research Focus

Our recent research is focused on using computational modeling, in particular machine learning, to advance understanding of how intelligence and cognition emerge in the brain. In this domain, the machine learning and AI lens complements the traditional views of experimental science and mechanistic models. Our research group develops machine learning methodology together with theory that can help explain the behavior of the underlying algorithms.

More info

Publications

Convergence and alignment of gradient descent with random backpropagation weights

We prove convergence of the "feedback alignment" algorithm for two-layer networks, proposed as a biologically plausible alternative to gradient descent. We also show that alignment of the weights requires regularization.

Recent publications