• Interdisciplinary AI
  • Computer (Vision) Scientist
  • Software Engineer

Hi, this is Jason speaking|

who .

NEW! Interactive Classification for Deep Learning Interpretation accepted to CVPR 2018 (Demo).
NEW! Featured in Georgia Tech College of Computing News for work presented at CVPR 2018.

About Me

headshot

LinkedIn
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  research under Professor Polo Chau & James Hays. Before that, I earned my dual degree in Computer Science & Business Administration at the University of Southern California. My research focus primarily deals with the interpretability of deep learning phenomenas such as Adversarial Attacks, and in explaining their robustness via visualization and other quantitative means. I also work closely with object detection, tracking, neural architecture search, GANs & inpainting.

A maker today, I believe the best way to improve our world is to add value through innovation and global leadership. From attending 30 hackathons and a dozen tech conferences, to working at large corporations i.e. Apple to Spotify, I believe that the key to success lies in self-motivation and unwavering grit.

In my spare time, I enjoy hiking outdoors, exploring new places, reading inspirational bios/stories, and indulging in history. On a daily basis, I immerse myself in HackerNews and a spectrum of tech feeds.


#HackathonWinners
                                                      

2018

CVPR: Visualizing Classifers

Interactive Deep Learning Classification

PGM - Regions of Interest

Graphical Models in Saliency Detection

AdVis.js

Exploring Adversarial Attacks

2017

OSx86 Hackintosh (education)

macOS Sierra on Skylake + Z170

Artemis

Food-tracking IoT Intelligence

Keplerian VR

Immersive team collaboration in VR

Stockpedia

Data Driven Stock Analytics

Cue: Bias-busting VR

Exposing unconscious bias by Q&A

2016

SoundScape

Location Aware Speakers

Taskify

Predictive Task Scheduler

SudoMap

Never miss a beat

OlympicBPM

Biosensor in Olympics

FooedBar

VR Food Experience

Jane Street etc.

High Freq. Electronic Trading Challenge

VeriRoute

Rating routes by driver safety metrics

Speed of Sound

NLP: Summarize & review on the fly

2015

ShÜber

Native Android App

TweeTINT

Sentiment Analysis Webapp

Data Visualization

D3.js rendering

Pebble

Location-based watchface

Groupazza

Study group & chatrooms

Anonimo

Real time Q&A in the classroom

AceCode

The App for every developer

Innovation Insight

Computer vision for the blind

Jola!

Android personal assistant

Vectorizor

Pencil sketch to SVG


Blog

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

Visit Blog