Welcome!

My name is Zi WANG, and I am a PhD candidate at EECS, MIT and part of LIS at CSAIL. I am working with Prof. Stefanie Jegelka, Prof. Leslie Pack Kaelbling, and Prof. Tomás Lozano-Pérez on exciting projects about active learning and Bayesian optimization with an ultimate goal of embedding the learning abilities in robots that help with household chores.

More about me

News

    Recent news


  • [Jun 29, 2018] Paper accepted at IROS, 2018.
  • [Dec 22, 2017] Paper accepted at AISTATS, 2018.
  • [May 15, 2017] Two papers have been accepted at ICML, 2017.
  • [Jan 15, 2017] Our paper Focused Model-Learning and Planning for Non-Gaussian Continuous State-Action Systems has been accepted at ICRA, 2017.
  • [Feb 2, 2016] I got my Master of Science in Electrical Engineering and Computer Science.
  • [Dec 23, 2015] Our paper Optimization as Estimation with Gaussian Processes in Bandit Settings has been accepted as oral presentation at AISTATS, 2016.
  • [Aug 10, 2015] I am TAing 6.883 Advanced Machine Learning  —  Learning with Discrete Structure this semester.
  • Recent talks/presentations


  • [Jul 3, 2018] Bayesian Optimization Guided by Max-values @ ISMP 2018, Bordeaux, France. [slides]
  • [Mar 9, 2018] Integrating model learning and TAMP @ LIS, MIT, Cambridge, MA. [slides available upon request]
  • [Nov 30, 2017] Bayesian Optimization and How to Scale It Up @ Computer Science Colloquium, University of Southern California, Los Angeles, CA. [slides]
  • [Aug 9, 2017] Max-value Entropy Search for Efficient Bayesian Optimization @ ICML 2017, Sydney, Australia. [slides] [video]
  • [Aug 7, 2017] Batched High-dimensional Bayesian Optimization via Structural Kernel Learning @ ICML 2017, Sydney, Australia. [slides] [video]
  • [Jun 20, 2017] (remote talk) Scaling up Bayesian Optimization with Ensembles @ DeepMind, London, UK. [slides]
  • [Jun 16, 2017] Focused Model-Learning and Planning for Non-Gaussian Continuous State-Action Systems @ Uber ATG, Pittsburgh, PA.
  • [Jun 7, 2017] Focused Model-Learning and Planning for Non-Gaussian Continuous State-Action Systems @ The Manipulation Lab, CMU Robotics Institute, Pittsburgh, PA.
  • [May 31, 2017] Focused Model-Learning and Planning for Non-Gaussian Continuous State-Action Systems @ ICRA 2017, Singapore. [poster][slides]
  • [Apr 28, 2017] Challenges in Long-horizon Planning and How to Learn a "Few-shot" Precondition Generator @ LIS, MIT, Cambridge, MA. [slides available upon request]
  • [May 9, 2016] Optimization as Estimation with Gaussian Processes in Bandit Settings @ AISTATS 2016, Cadiz, Spain. [slides]
  • [May 2, 2016] Optimization as Estimation with Gaussian Processes in Bandit Settings @ Machine Learning Tea, MIT, Cambridge, MA.

Research Interest

Learning and planning, and how they interact.


Education

Massachusetts Institute of Technology

2014 - present

Ph.D. student in Electrical Engineering and Computer Science
Advisors: Prof. Stefanie Jegelka, Prof. Leslie Kaelbling, and Prof. Tomás Lozano-Pérez

Massachusetts Institute of Technology

2014 - 2016

S.M. in Electrical Engineering and Computer Science
Master thesis: Optimization as Estimation with Gaussian Process Bandits
Thesis Supervisors: Prof. Stefanie Jegelka and Prof. Leslie Kaelbling

Tsinghua University

2010 - 2014

B.Eng. in Computer Science and Technology
Undergraduate thesis: Fast Dropout Training for Deep Neural Networks (in Chinese)
Thesis advisors: Prof. Fei Sha and Prof. Jun Zhu


Publications

2018


Active model learning and diverse action sampling for task and motion planning

Zi Wang, Caelan Reed Garrett, Leslie Pack Kaelbling, Tomás Lozano-Pérez

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018

Abstract PDF arXiv Code Project page BibTex

Batched Large-scale Bayesian Optimization in High-dimensional Spaces

Zi Wang, Clement Gehring, Pushmeet Kohli, Stefanie Jegelka

International Conference on Artificial Intelligence and Statistics (AISTATS), 2018

Abstract PDF arXiv Code BibTex

2017


Batched High-dimensional Bayesian Optimization via Structural Kernel Learning

Zi Wang, Chengtao Li, Stefanie Jegelka, Pushmeet Kohli

International Conference on Machine Learning (ICML), 2017

Abstract PDF arXiv Code Slides BibTex

Max-value Entropy Search for Efficient Bayesian Optimization

Zi Wang and Stefanie Jegelka

International Conference on Machine Learning (ICML), 2017

Abstract PDF arXiv Code Slides BibTex

Focused Model-Learning and Planning for Non-Gaussian Continuous State-Action Systems

Zi Wang, Stefanie Jegelka, Leslie Pack Kaelbling, Tomás Lozano-Pérez

IEEE Conference on Robotics and Automation (ICRA), 2017

Abstract PDF arXiv Project page Video Slides BibTex

2016


Optimization as Estimation with Gaussian Processes in Bandit Settings

Zi Wang, Bolei Zhou, and Stefanie Jegelka

International Conference on Artificial Intelligence and Statistics (AISTATS), 2016

Oral presentation (6% acceptance rate)

Abstract PDF arXiv Code Project page Slides BibTex

2014


Fast Learning with Noise in Deep Neural Nets

Zhiyun Lu*, Zi Wang*, and Fei Sha

(Spotlight Presentation) NIPS Workshop: Perturbations, Optimization, and Statistics, Quebec, Canada, 2014

Abstract PDF BibTex

Discriminative Non-Negative Matrix Factorization for Single-Channel Speech Separation

Zi Wang and Fei Sha

IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Florence, Italy, 2014

Abstract PDF Project page IEEE Poster BibTex

2013


Scalable Inference for Logistic-Normal Topic Models

Jianfei Chen, Jun Zhu, Zi Wang, Xun Zheng, and Bo Zhang

Advances in Neural Information Processing Systems (NIPS), Lake Tahoe, CA, 2013

Abstract PDF Project page NIPS Poster Code BibTex

Contact Me

ziw 'at' mit.edu

To family and friends who think computer science is about making websites and fixing your computer: yes, I am physically able to do these, but they are not intellectually challenging enough for me, as an AI major. Please consult the people who study networks and architecture on the 7th floor of Stata (JK!).