Alexander Amini

email: amini@mit.edu
room: MIT 32-376
CV // google scholar // bio

Bio

I am the co-founder and Chief Science Officer of Liquid AI, and a Research Affilliate at Massachusetts Institute of Technology (MIT). I completed my PhD (2022), Master of Science (2018), and Bachelor of Science (2017) in Computer Science at MIT, with a minor in Mathematics.

The objective of my research is to develop the science and engineering of autonomy and its applications to safe decision making for autonomous agents. My vision is a world with adaptive autonomous agents capable of learning to interact in complex, uncertain, and extreme scenarios, supporting people with cognitive and physical tasks. I have worked on learning end-to-end control (i.e., perception-to-actuation) of autonomous systems, formulating confidence of neural networks, mathematical modeling of human mobility, as well as building complex inertial refinement systems.

I am also the co-founder of Themis AI and the lead organizer and lecturer for MIT 6.S191: Introduction to Deep Learning, MIT's official introductory course on deep learning. In high school, I was awarded the first place Grand Prize at the EU Content for Young Scientists and BTYSTE with my project entitled: Tennis Sensor Data Analysis: An Automated System for Macro Motion Refinement.

Please click here for my curriculum vitae (CV) or a third-person bio.

News

Oct 2021	Paper: Two new papers accepted to NeurIPS 2021 on the topics of causual navigation models [paper] and sparse flows [paper]!
Apr 2021	Our paper on co-optimizing sensor placement and policy learning is accepted to RA-L and nominated for the Best Paper Award at RoboSoft 2021! [paper] [news]
Mar 2021	Two new papers using our evidential uncertainty algorithm have been accepted and published for improved robustness in molecular drug discovery [paper] and end-to-end autonomous driving [paper]!
Feb 2021	Awarded two grants with FinTech@CSAIL and MachineLearningApplications @CSAIL to mitigate algorithmic bias and uncertainty in financial time series modeling and clinical trial outcome prediction.
Jan 2021	Lead organizer and lecturer for MIT 6.S191: Introduction to Deep Learning, with over 700 registered MIT students (and over 30,000 registrations globally online). [link] [video]
Dec 2020	Awarded the JP Morgan Fellowship for the 2021-2022 academic year with a focus on robustness and uncertainty of learning-based systems!
Dec 2020	Our paper on liquid time-constant neural networks is accepted to AAAI with an Oral spotlight! [paper] [news]
Aug 2020	Two papers focused on uncertainty and robustness of ML! (1) published in Nature Machine Intelligence and (2) accepted at NeurIPS for presentation in Dec 2020.
Jun 2020	Awarded: the annual MIT Outstanding Mentor Award for Undergraduate Researchers to recognize an individual research mentor who has demonstrated exceptional guidance and teaching in a research setting. Thank you to my students!
May 2020	Paper: Learning end-to-end control policies from data-driven simulation has been accepted to RA-L with an invited presentation at ICRA 2020! Code coming soon. [paper]
Jan 2020	Lead organizer and lecturer for MIT 6.S191: Introduction to Deep Learning, with over 300 MIT registered students. [link] [video]
Oct 2019	Invited Talk and contributed papers at ICML 2019 Workshop for Autonomous Driving, and Reinforcement Learning in Real Life.
Mar 2019	Our paper Variational End-to-End Navigation and Localization has been nominated for the Best Paper Award at ICRA 2019 (top 1% of submissions)
Jan 2019	Paper: Variational End-to-End Navigation and Localization has been accepted to ICRA 2019!
Jan 2019	Lecturer for the 2nd straight year of MIT 6.S191: Introduction to Deep Learning, MIT's official course on deep learning applications and foundations. [link] [video]
Dec 2018	Paper: Uncovering and Mitigating Algorithmic Bias through Learned Latent Structure has been accepted to AAAI/ACM AIES 2019 [paper]
Oct 2018	Invited Talk at IROS in Madrid, Spain: Variational Autoencoder for End-to-End Control of Autonomous Driving with Novelty Detection and Training De-biasing
Jun 2018	Paper: Variational Autoencoder for End-to-End Control of Autonomous Driving with Novelty Detection and Training De-biasing has been accepted to IROS 2018 [paper]
Jun 2018	Graduation: Master of Science (MS) from MIT in EECS. Thesis: Robust Learning for End-to-End Autonomous Driving [thesis]
Mar 2018	Invited Talk at NVIDIA's GTC in San Jose, California: Learning Steering Bounds for Parallel Autonomy [talk]
Jan 2018	Lecturer for MIT 6.S191: Introduction to Deep Learning, MIT's official course on deep learning applications and foundations. [link] [video]
Dec 2017	Invited Talk at the NIPS workshop on Bayesian Deep Learning (12% acceptance rate).
Nov 2017	Travel Award to present our paper, Spatial Uncertainty Sampling for End-to-End Control, at NIPS Bayesian Deep Learning (8% acceptance rate). [paper]
Jun 2017	Summer Internship starting my summer internship with NVIDIA's end-to-end self driving car team.
Jun 2017	NSF Fellowship. Awarded the National Science Foundation Graduate Fellowship (10% acceptance rate).
Jun 2017	Graduation: Bachelor of Science (BS) from MIT in EECS with a minor in Mathematics and concentration in Economics.

Load More