PhD · Neural Computation & ML · Carnegie Mellon University

Hi, I’m Yuki Minai

I am an ML & neurotech engineer, developing ML-driven neuroscience technology to augment human capacity. Currently, I am a PhD student in Neural Computation and Machine Learning at Carnegie Mellon University, working with Prof. Byron Yu and Prof. Matthew Smith.

Yuki Minai
CMU
PhD program · Pittsburgh
12+
Technical blog posts
BCI · Closed-loop stimulation
RL · DL · Signal Processing
View all →

My current research focus is the development of a closed-loop brain stimulation framework to induce desired brain states. I combine a wide range of ML techniques (such as reinforcement learning, deep learning, and latent variable models) to develop a brain stimulation framework that adaptively update brain stimulation parameters in real-time.

In the long term, I am passionate about making sociatal impacts by developing ML-driven neuroscience technologies to improve our lives. While the current applications of neuroscience technologies are largely within the medical field, I believe in a future where these innovations enhance human capacity and transform our lives for the better in diverse ways. My current research project is my first step toward realizing this vision.

$ Research project spotlight
OMiSO — Online MicroStimulation Optimization NeurIPS 2025
To improve the accuracy by which one can manipulate neural activity, it is important to (1) take into account the pre-stimulation brain state, which can influence the brain’s response to stimulation, and (2) adaptively update stimulation parameters over time to compensate for changes in the brain’s response to stimulation. In this work, we propose OMiSO, a brain stimulation framework that leverages brain state information to adaptively update stimulation parameters that can drive neural population activity toward specified states.
Closed-loop optimization Reinforcement learning Brain-computer interface NeurIPS 2025
MiSO — MicroStimulation Optimization NeurIPS 2024
Brain stimulation has the potential to create desired neural population activity states. However, it is challenging to search the large space of stimulation parameters, for example, selecting which subset of electrodes to be used for stimulation. To address this challenge, we develop MiSO, a closed-loop stimulation framework to drive neural population activity toward specified states by optimizing over a large stimulation parameter space.
Closed-loop optimization Deep learning Brain-computer interface NeurIPS 2024
yukiminai.com · CMU PhD · ML & Neuroscience Built with care · Medium · GitHub
Scroll to Top