chandan ⃝ cs1@berkeley.

edu
⃝ 571-315-5748
⃝
⃝
csinva

csinva.github.io
⃝
⃝
singh
neural network models neurobiology
education visual neuroscience cognitive science
phd | computer science
uc berkeley | 2017-present
• research: interpretable ml,
computational neuroscience
• advisor: bin yu
• collaborators: jack gallant
• gpa: 3.92

bs | computer science &

coursework
computation
machine learning
computer vision
structure learning
algorithms
artificial intelligence
deep learning
learning theory
deep learning in
graphics theory of
computation data
structures software
dev. I & II
stat/math
statistical models
probability
statistics
optimization
linear
algebra info
theory real
analysis
linear models
stochastic processes
chaos theory I & II
multivariate calculus
discrete mathematics
differential equations
abstract algebra
neuroscience
neural coding
1/2
 • implemented crfs for segmentation post-processing
experience
berkeley b. yu research uva y. qi research lab | ml researcher
lab | ml researcher fall 2016 – spring 2017
fall 2017 - present • developed novel weighted-ℓ1, multi-task gaussian graphical model
• investigated methods to • analyzed large-scale functional brain connectivity with graphical models
interpret deep learning models
• developed machine-learning hhmi s. turaga research lab | ml researcher
algorithms to model neural summer 2015, winter 2015, summer 2016
data • extended novel watershed algorithms for neural image segmentation
• developed statistical methods • contributed to development of novel 3d unet cnn architecture with malis loss
to learn from small data • distributed mllib random forest over compute cluster with apache spark
pacmed ai | interpretable uva w. levy research lab | comp. neuroscience researcher
ml intern fall 2014 - fall 2016
summer 2019 • simulated detailed biophysical neurons to understand neural computation
• will develop techniques to • analyzed energy efficiency, noise, and variability of neural computation
interpret machine-learning via stochastic sodium-channel gating
models for healthcare
hhmi scientific computing | research intern
facebook | computer vision summer 2014
intern • examined effects of back-propagating action potentials by simulating
summer 2017 intracellular neural firing with detailed biophysical models
• improved deep learning • simulated extracellular recording from neurons and measured noise
models for semantic • made detailed visualizations of action potential firing
segmentation of satellite
imagery research innovations inc. | web dev / android intern
• investigated autoencoders for summer 2013 - spring 2014
unsupervised layer-wise • developed web application to simultaneously coordinate different tasks
pretraining • developed android app to increase data storage capacity of qr codes
web/mobile
teaching basic languages • javascript • django android • jekyll • mapping apis
berkeley | student instructor
summer 2018
• cs 189/289: machine learning projects (non-research)
• lectures to class of 80+ students  hummingbird tracking with opencv  
news balancer django app 
notes, blog, & slides 
skills
languages
experienced funding awards
python • java • matlab utokyo neurocomputing workshop travel award (2019)
proficient pdsoros fellowship finalist (2019) vidya shelat fund award (2016) rodman
r • c++ • c • mathematica scholar (2014-2017)

machine learning
frameworks
pytorch • tensorflow • scikit-
learn keras • mllib • caffe
algorithms
cnns • graphical models • rfs

general
languages
english • spanish • hindi
software
LATEX• photoshop • NEURON
os
linux • mac • windows
ides
jupyter • intellij • eclipse • vim

2/2
 ∗ ∗
papers / posters • murdoch , singh , kumbier, abbasi-asl, & yu, 2018: “interpretable machine
learning: definitions, methods, and applications” 
published/accepted
∗ ∗ • levy lab: “neural computation at the thermal limit”  
• singh , murdoch , & yu,
talks
2019: “hierarchical
interpretations for neural • singh, 2017: “a novel machine-learning algorithm for uncovering brain
network predictions” iclr  connections underlying autism” uva undergraduate research & design symposium,
 winner in design category 
∗ ∗ • singh, 2017: “uncovering brain connections underlying autism via graphical
• funke , tschopp , grisaitis,
sheridan, singh, saalfeld, & models”
turaga, 2018: “large scale tom tom founder’s machine learning conference 
image segmentation with • singh, 2017: “complexity leads to simplicity: investigating neural linearization
structured-loss-based deep via biophysical simulations” uva undergraduate research & design symposium
learning for connectome semifinalist in research category (1 of 6 undergraduates) 
reconstruction” tpami  posters
∗ ∗
• morel, singh, & levy, 2018: • singh , murdoch , & yu, 2018: “interpretable machine learning with
“linearized synaptic integration at applications to neuroscience” utokyo neurcomputing workshop 2019
no extra cost” journal of ∗ ∗
• singh , murdoch , & yu, 2018: “hierarchical interpretations for neural
computational neuroscience  network predictions” berkeley bair workshop fall 2018, iclr 2019 
• singh, wang, & qi, 2017: “a • singh, hewitt, & turaga, 2015: “optimizing random forest image
weighted-ℓ1, multi-task graphical
segmentation for connectomics” janelia undergraduate scholar poster session 
model with applications to
heterogeneous brain
connectivity” nips 2017 amlicd
workshop   awards
• singh & levy, 2017: uva rader research award 2017
“a consensus layer uva undergraduate research symposium winner 2017
V pyramidal neuron raven honor society 2016-2017
can sustain icpc regional qualification 2014-2016
interpulse-interval 1st place microsoft code competition 2016
coding” plos one.  3rd place google games uva 2017
 2nd place apt puzzle competition 2017
under review intermediate honors 2016
dean’s list 2014-2017

outside activities
im basketball, soccer, frisbee 2015-2018
apda, pf debate 2010-2017
indian student association 2014-2017
madison house volunteering (computer literacy) 2014-2017
chinmaya mission volunteering 2010-2014

3/2