Jeremy Irvin


Computer Science Master's

Stanford









Education
Stanford
Computer Science Master's Degree
Sep 2016 - June 2018
Stanford Master's

Completing my Master's degree in Computer Science with a specialization in AI.

4.00 / 4.00 GPA

Coursework thus far:
CS229
CS221
CS228
CS224N
STATS315A

UCSB
Double B.S. Math and CS
Sep 2012 - June 2016
UCSB Double B.S.

Obtained a B.S. in both Math and Computer Science in the College of Creative Studies at UCSB.

3.97 / 4.00 GPA

Regents Scholar, Graduated with Highest Honors

Recent Projects
Recurrent Neural Networks with Attention for Genre Classification
Fall 2016
Recurrent Neural Networks with Attention for Genre Classification
Implemented RNN’s and LSTM’s for automatic genre classification of songs using audio spectrograms.

All code written in Python using TensorFlow.

Achieved results comparable to state-of-the-art which uses hand-crafted features.

Paper
Poster
Implementation
Convolutional Neural Networks for 3D MNIST Image Classification
Fall 2016
Convolutional Neural Networks for 3D MNIST Image Classification
Implemented 3D CNN's for classification of a 3D MNIST dataset.

All code written in Python using TensorFlow.

Most of the models beat all of the baselines.

Paper
Poster
Implementation
Deep Learning
Bootcamp
Fall 2016
Deep Learning Bootcamp
Participated in a selective (5-student) deep learning bootcamp under Professor Andrew Ng and his PhD students.
Implemented the following three papers (using Python with TensorFlow):

Neural Machine Translation with Attention
Implementation

Variational Autoencoder
Implementation

Zero Shot Translation
Implementation





Internships
Microsoft
Summer 2016
Microsoft
Interned on the Bing Predicts team, a team of machine learning experts and data scientists from Microsoft Research and Bing, making predictions on popular events in entertainment, sports, politics, and more.

Shipped predictions for the MTV Video Music Awards, getting 6/9 correct in the top 2. Wrote the model in Python using MART Gradient Boosting.

Additionally implemented LSTM's for time series forecasting using Keras and CNTK.
Microsoft
Summer 2015
Microsoft
Interned on the Satori (Knowledge Graph) team under Bing.

Developed an algorithm to detect subtle entity relations in an immense ontology and rank them by novelty.

Wrote C# and internal query language as part of an R&D ML pipeline.

Increased run-time by two orders of magnitude, allowing for efficient discovery of the relations.
Silicon Valley Education Foundation
Summer 2013
SVEF
Worked as part of 'Stepping Up To Algebra', an intervention program to help students complete Algebra I and increase student aspirations to attend college.

Instructed and tutored a large group of at risk students in basic algebra.

Helped plan and conduct the final cumulative Parent-Teacher college preparation meeting.

Research, Teaching, and Skills
Linguistics Research
Aug 2015 - June 2016
Linguistics Research
Work with Professor Moscoso at UCSB.

Using dynamical systems and causal modeling techniques to understand the underlying complex nonlinear patterns of language development in children.

First Author Paper in Cognitive Science
Conference Poster
Conference Talk

Second Author Paper in ACL

Final Thesis
Computer Learning Co-Lecturer
Winter 2016
Computer Learning Co-Lecturer

Co-lead a course on Machine Learning, NLP, and Deep Learning.

Created more than 250 slides of lectures from notes from various sources.


Skills
Ongoing
Skills

Languages:
Python (5 years),
C (4 years),
R, C++, C#, SQL

Notable Tools/Frameworks:
TensorFlow, scikit-learn, Scrapy, Selenium, Git, LaTeX