Lanlan Liu


Contact: llanlan [at] umich [dot] edu

I recently graduated from Computer Science and Engineering at University of Michigan, advised by Prof. Jia DengI visited Princeton University from November 2018 to June 2020. I obtained my bachelor degree from University of Science and Technology of ChinaMy research interest is focused on reducing human designs in deep learning and computer vision through dynamic inference, unified loss framework, architecture search, and synthetic data generation. I did internship in Summer and Fall 2018 at Google Cloud AI Research, advised by Dr. Tomas Pfister and Dr. Jia Li, and with AWS Rekognition and Video team in summer 2019. After graduation, I joined Facebook as a full-time in summer 2020.

Education:

09/2015–06/2020 Ph.D., Computer Science and Engineering
Computer Science and EngineeringUniversity of Michigan, Ann Arbor
GPA:                      4.0/4.0 
Advisor:                Prof. Jia Deng 

08/2011–06/2015 B.E., Computer Science and Technology
School for the Gifted YoungUniversity of Science and Technology of China
-The Talent Class for Computer and Information Science
GPA:                      3.89/4.30
Ranking:               2/99

Publications:

Dynamic Deep Neural Networks: optimizing Accuracy-Efficiency Trade-offs by Selective Execution
Lanlan Liu
, Jia Deng 
AAAI, 2018  [link

Generative Modeling for Small-Data Object Detection
Lanlan Liu, Michael Muelly, Jia Deng, Tomas Pfister, Li-Jia Li
ICCV, 2019 (Oral Presentation) [link]

A Unified Framework of Surrogate Loss by Refactoring and Interpolation
Lanlan Liu, Mingzhe Wang, Jia Deng 
ECCV, 2020 (Spotlight Presentation) [link]

Dynamic Grown Generative Adversarial Networks
Lanlan Liu, Yuting Zhang, Jia Deng, Stefano Soattos
AAAI, 2021



Professional Activities:

Reviewer:
    - IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
    - International Conference on Computer Vision (ICCV)
    - Winter Conference on Applications of Computer Vision (WACV)
    - 
Asian Conference on Machine Learning (ACML)
    - AAAI Conference on Artificial Intelligence (AAAI)
    - Conference on Computer Vision and Pattern Recognition (CVPR)

Talks:
    - Dynamic Deep Neural Networks. CVPR 2017 WiCV Workshop
 
    
- Dynamic Deep Neural Networks. NeurIPS 2018 WiML Workshop 


Intern Experience:

12/2018-06/2020 Visiting StudentPrinceton Universitysupervised by Prof. Jia Deng 
06/2019-09/2019 Applied Scientist InternAWS Rekognition and Video, supervised by Dr. Yuying Zhang
05/2018-04/2019 Student ResearcherGoogle Cloud AI, supervised by Dr. Tomas Pfister And Dr. Jia Li 
09/2014-01/2015 Research InternMicrosoft Research Asiasupervised by Dr. Tie-Yan Liu and Dr. Taifeng Wang
07/2014-09/2014 Visiting StudentUCLA CSST Programsupervised by Prof. Alan L. Yuille 
07/2013-08/2013 Research InternMicrosoft Research Asia, supervised by Dr. Tie-Yan Liu and Dr. Taifeng Wang

Activities and Services:

2017-2018 President of USTC Alumni Association in Great Detroit Area
2016-2018 Corporate Chair of Ensemble of Computer Science and Engineering Ladies @UofM
2016 Oct. Volunteer and attendee of Grace Hopper Celebration 2016
2012-2013 Vice Chair of Student Union of School for the Gifted Young @USTC
2011-2013 Volunteer at Volunteer’s Team for Alumni @USTC

Selected Honors:

06/2017 Rackham Conference Travel Grant, University of Michigan
08/2016 Grace Hopper Celebration Travel Grant, University of Michigan
07/2014 Google Anita Borg Scholarship ( Top 3% )
06/2014 Microsoft-IEEE Young Fellowship ( Top 2% )
10/2011 Gifted New Student Award, USTC Oversea Alumni Foundation ( Top 0.2% )
10/2013 Award of Excellent Intern, Stars of Tomorrow Program, MSRA ( Top 15% )