• Human-centered AI
  • Computer (Vision) Scientist
  • Software Engineer

Hi, this is Jason Lin|

who .

 NEW!  Deep Learning & Adversarial Attack open source demos AdVis and DeepVis are live now.
 NEW!  Featured in Georgia Tech College of Computing News for demo presented at CVPR 2018.

About Me


Jason Lin

Hi! I'm currently pursuing my M.S. in Computer Science at Georgia Institute of Technology, specializing in Machine Learning and Computer Vision . Before that, I earned my dual degree in Computer Science & Business Administration at the University of Southern California.

My research interests deal with the interpretability of neural networks such as Adversarial Attacks, and explaining their robustness via visualization and other quantitative means. I also work closely with object detection & tracking, image restoration, and efficient neural networks for few-shot learning.

A maker today, I believe the best way to improve our world is to add value through innovation, unwavering grit, and global leadership. In my spare time, I enjoy hiking outdoors, exploring new places, reading inspirational bios/stories, and indulging in history.



Interactive Classification for Deep Learning Interpretation
Angel Cabrera, Fred Hohman, Jason Lin, Duen Horng Chau.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018 (Demo)

arXiv. 2018.


AdVis: Visualizing and Attributing ML Attacks to Adversarial Inputs
Jason Lin, Dilara Soylu.
Advanced Computer Vision advised by James Hays
poster. 2018.


Detecting Graphical Regions of Interest with Gaussian Process Bayesian Optimization
Jason Lin, Hakki Mert Torun, Weihua Zhu, Jingjing Pan.
Probabilistic Graphical Models advised by Faramarz Fekri
slides. 2018.


Two-stage vs. One-stage detectors: State-of-the-art Object Detection and Tracking
Jason Lin.
In progress.


Network Virtualization: Distributed Computing and a more secure Internet
Jason Lin.
Enterprise Network Design, cybersecurity specialization (USC)
paper. 2016.


Quantum Computing and its effects on Deciphering Public Key Cryptosystems
Jason Lin.
Ethical Hacking and Systems Defense, cybersecurity specialization (USC)
paper. 2015.

Research interests: computer vision, deep learning, human-computer interaction  
Contact: jlin401 [at] gatech [dot] edu



CVPR: Visualizing Classifers

Interactive Deep Learning Classification


Exploring Adversarial Attacks

PGM - Regions of Interest

Graphical Models in Saliency Detection


Graduate Computer Vision

SIFT, Calibration, Scene Understanding

Cue: Bias-busting VR

Exposing unconscious bias by Q&A


Food-tracking IoT Intelligence


Data Driven Stock Analytics

Keplerian VR

Immersive team collaboration in VR



Location Aware Speakers


Biosensor in Olympics


Never miss a beat

Speed of Sound

NLP: Summarize & review on the fly


VR Food Experience


Innovation Insight

Computer vision for the blind


Study group & chatrooms


Android personal assistant


Pencil sketch to SVG


Sentiment Analysis Webapp

Data Visualization

D3.js rendering



Like what you see? There's always more. Somewhere.

Visit Blog