Neurons – these elegant, plastic microchips of the brain and their synaptic networks enable us to learn, adapt, envision new ideas, build brain-inspired machines, and create art and science. Understanding how the brain uses its structural and physiological components to achieve sophisticated cognitive abilities requires computational models that integrate these features step-by-step, ultimately revealing “what makes us human.”
Selected Lab Projects
- Connectomics-based NEURON Simulations: We leverage “big data” from dense reconstructions (Connectomics Project) to create realistic models of cortical neurons and networks. Using compartmental models, machine learning, and graph theory, we explore how neurons’ geometry and network connectivity shape the I/O properties of neurons and circuits.
- NeurAI: This project develops deep neural network analogues to human cortical neurons, using experimental data from brain tissue obtained during epilepsy or tumor surgeries. The goal is to uncover what gives human neurons and circuits their unique computational power.
- Energy Efficiency of the Brain: Our brain performs complex cognitive tasks using just 20 watts – equivalent to a light bulb – compared to megawatts required by supercomputers for similar tasks. We investigate how analog computations in dendrites and synapses enable this efficiency, with the aim of developing energy-efficient, neuromorphic systems.
- Frontiers for young minds. We believe in preparing kids for the fast-changing world – via the open access FYM articles, written by scientists and reviewed by kids
Our work is supported by grants from the National Institutes of Health (NIH) and the Office of Naval Research (ONR).
