Lu Mi

Lu Mi 米璐

I am currently a Shanahan Foundation Fellow at the Allen Institute, and postdoctoral researcher at the University of Washington, Seattle, where I work closely with Dr. Uygar Sümbül, Prof. Matthew Golub, Prof. Edgar Walker, and Prof. Eric Shea-Brown. Prior to that, I received my Ph.D. at MIT CSAIL in 2022, where I studied in Computational Connectomics Group. I was fortunate to be advised by Prof. Nir Shavit at MIT, and co-advised by Prof. Aravinthan D.T. Samuel and Prof. Jeff W. Lichtman at Harvard Center for Brain Science. Meanwhile, I was a visiting researcher at HHMI Janelia Research Campus working with Dr. Srinivas C. Turaga. I also held research internship at Google and Waymo.

I received my M.S. degree at MIT EECS in 2019, and B.S. degree at Tsinghua in 2017.

Email / Google Scholar / Twitter / CV

Research

My research interest is NeuroAI, which studies the intersection of natural intelligence and artificial intelligence. During my PhD, I developed advanced deep learning tools for fast and scalable automatic connectomics pipelines to discover the brain, and linking the anatomical structure and neural activity with whole-brain modeling. In the current collaboration with investigators at the Allen Institute and the University of Washington, I am developing interpretable deep learning approaches including probabilistic modeling and representation learning for biological and artificial systems. My research topics broadly include
(1) Developing fast and scalable automatic pipelines to discover the brain;
(2) Modeling brain with multi-modal neural data;
(3) Understanding the robustness and efficiency of coding, computation and learning in biological and artificial systems;
(4) Building brain-inspired AI frameworks.

I am currently on academia job market in fall/winter 2023. Meanwhile, I'm looking for collaborations with motivated students who are passionate about NeuroAI. Drop me an email if you are interested!

News

[09/23] Our work on NeuPRINT: Time-Invariant Neural Representation Learning was accepted at NeurIPS 2023.
[03/23] The Shanahan Undergraduate and Postbac Fellowships are open to apply.
[02/23] I gave a guest lecture on Biophysics VAEs in Deep Learning for Neuroscience course (CSE 599N) at University of Washington, Seattle.
[01/23] Our work on Xray2EM: Cross-Modality Image Reconstruction was accepted at ISBI 2023.
[11/22] Our organized challenge on Xray Projectomic Reconstruction was accepted at ISBI 2023.
[09/22] I gave a short talk on How to Link Multi-Modal Neural Data with Deep Learning at NeuroAI in Seattle 2022.
[09/22] I joined Allen Institute as Shanahan Foundation Fellow, and University of Washington, Seattle as postdoctoral researcher.
[08/22] Ph.D. Thesis defended!
[08/22] I received the 2022 Shanahan Foundation Fellowship.
[08/22] I was selected into 2022 Rising Stars in EECS.
[04/22] I gave a talk on Deep Learning Tools for Next-Generation Connectomics at Allen Institute.
[01/22] Our work on Connectome-Constrained Modeling was accepted at ICLR 2022.
[01/22] I gave a talk on Connectome-Constrained Modeling in Computational Neuroscience Seminar at Flatrion Institute.
[12/21] Our work on Training-Free Uncertainty Estimation was accepted at AAAI 2022.
[06/21] I joined Google Research as a research intern.
[06/21] I received the 2021 MathWorks Fellowship.
[02/21] Our work on HD Maps Generation was accepted at CVPR 2021.
[01/21] Our work on Connectome-Constrained Modeling was accepted at Cosyne 2021.
[09/20] I received the NIH Awards at MICCAI 2020.

Selected Publications

* indicates equal contribution
	Learning Time-Invariant Representations for Individual Neurons from Population Dynamics Lu Mi, Trung Le, Tianxing He, Eli Shlizerman, Uygar Sümbül NeurIPS 2023 We use implicit dynamical models to learn time-invariant representation for individual neurons from population dynamics and enables mapping functional activity to transcriptomic identity.
	Connectome-constrained Latent Variable Model of Whole-Brain Neural Activity Lu Mi, Richard Xu, Sridhama Prakhya, Albert Lin, Nir Shavit, Aravinthan D.T. Samuel, Srinivas C. Turaga ICLR 2022 We use stochastic threshold linear dynamics to model the whole-brain neural network, using connectome constraint on the neural activity. pdf / code
	Training-Free Uncertainty Estimation for Dense Regression: Sensitivity as a Surrogate Lu Mi, Hao Wang, Yonglong Tian, Hao He, Nir Shavit AAAI 2022 ICML 2021 Workshop on Uncertainty & Robustness in Deep Learning We perform a systematic exploration into training-free uncertainty estimation for dense regression, an unrecognized yet important problem, and provide a theoretical construction justifying such estimations. arxiv / video / slides / poster
	HDMapGen: A Hierarchical Graph Generative Model of High Definition Maps Lu Mi, Hang Zhao, Charlie Nash, Xiaohan Jin, Jiyang Gao, Chen Sun, Cordelia Schmid, Nir Shavit, Yuning Chai, Dragomir Anguelov CVPR 2021 We propose HDMapGen, a hierarchical graph generative model capable of producing high-quality and diverse HD maps. pdf / video / slides / poster
	Learning Guided Electron Microscopy with Active Acquisition Lu Mi, Hao Wang, Yaron Meirovitch, Richard Schalek, Srinivas C. Turaga, Jeff W. Lichtman, Aravinthan D. T. Samuel, Nir Shavit MICCAI 2020 We show how to use deep learning to accelerate and optimize single-beam SEM acquisition of images. arxiv
	Cross-Classification Clustering: An Efficient Multi-Object Tracking Technique for 3-D Instance Segmentation in Connectomics Yaron Meirovitch, Lu Mi*, Hayk Saribekyan, Alexander Matveev, David Rolnick, Nir Shavit CVPR 2019* We introduce cross-classification clustering (3C), a technique that simultaneously tracks complex, interrelated objects in an image stack. pdf

Professional Services

Reviewer for NeurIPS, ICLR, CVPR, ECCV, ICCV, AAAI.
Guest lecture on Biophysics VAEs, Deep Learning for Neuroscience, CSE 599N, University of Washington, Seattle, 2023.
Talk on How to Link Multi-Modal Neural Data with Deep Learning, NeuroAI in Seattle, 2022.
Talk on Deep Learning Tools for Next-Generation Connectomics, Allen Institute, 2022.
Talk on Connectome-Constrained Modeling, Computational Neuroscience Seminar, Flatrion Institute, 2022.
Talk on Connectome-Constrained Modeling, CVML meeting, HHMI Janelia Research Campus, 2021.
Talk on Cross-Classification Clustering Segmentation, Machine Learning & Biology NSF Workshop, 2019.
TA for MIT 6.555 (Biomedical Signal and Image Processing) by Julie Greenberg, 2019.

Honors and Awards

Shanahan Foundation Fellowship, Allen Institute & University of Washington, Seattle, 2022
Risng Stars in EECS, 2022
MathWorks Fellowship, MIT EECS, 2021
NIH Awards, MICCAI, 2020
Grass Instruments Co. Fellowship, MIT EECS, 2017
Tang Lixing Fellowship, Tsinghua (30 in 3000), 2017
Best Paper Awards, Optofluidics, 2016
1st Prize (Meritorious Winner) COMAP's Mathematical Contest in Modeling (MCM), 2016

The design and code of this website is adapted from Jon Barron's site.