Developer Project

Nikhil Joshi

Physicist, Data-scientist, Software developer

*.*.*****@*****.***

Summary

I am a data scientist, with training in experimental physics, currently working in computational neuroscience.

My expertise and experience include

1. Statistics and data analysis -

Predictive modeling, Parameter estimation, linear discriminators and optimization

2. Machine learning -

Neural Networks, SVMs, Genetic algorithms (GA),

3. Programming

algorithms, and data structures

4. Languages (in order of fluency)

C++, C, Python, Matlab, R, Objective-C, Javascript

Experience

Postdoctoral Fellow at Swiss Federal Institute of Technology

April 2013 - Present (10 months)

Blue Brain Project (Human brain Project)

Postdoctoral Fellow at California Institute of Technology

March 2011 - March 2013 (2 years 1 month)

We evolved intelligent brains-like networks (represented by Mario in the accompanied videos) made of

twelve binary nodes and the connectivity between these nodes fully encoded in their respective genome,

which could solve two-dimensional mazes intelligently.

We found the minimal network complexity increases with the agent fitness, providing evidence that evolved

complexity is an admixture of necessity and chance.

- with Prof. Christof Koch (Caltech) and Prof. Giulio Tononi (UW Madison)

Visiting Fellow at ETH Zurich

September 2010 - December 2010 (4 months)

Worked with the BISON group in the Control Systems Lab of Department of Information Technology and

Electical Engineering

Graduate student at The Belle Collaboration

March 2005 - September 2010 (5 years 7 months)

Page1

Graduate Student at Tata Institute of Fundamental Research

August 2003 - September 2010 (7 years 2 months)

Study of CP-violation using the B-meson system at the KEK electron-positron collider in Japan

Visitor at The Institute of Mathematical Sciences, Chennai

2003 - 2003 (less than a year)

Loop Quantum Gravity

Master of Science at Indian Institute of Technology, Kanpur

2001 - 2003 (2 years)

Certifications

Machine Learning

Coursera December 2013

Volunteer Experience

Organizing committee volunteer at TEDx Caltech (2012-13)

January 2013 - January 2013 (1 month)

Grand Award Judge at LA County Science Fair

March 2012 - Present (1 year 11 months)

Grand Award Judge at Intel International Science and Engineering Fair

May 2011 - Present (2 years 9 months)

Projects

In-silico evolution of complex brains

Members:Nikhil Joshi, Prof. Christof Koch, Prof. Giulio Tononi

What is the relationship between the complexity and the fitness of evolved organisms, whether natural or

artificial? It has been asserted, primarily based on empirical data, that the complexity of plants and animals

increases as their fitness within a particular environment increases via evolution by natural selection. We

simulate the evolution of the brains of simple organisms living in a planar maze that they have to traverse as

rapidly as possible. Their connectome evolves over 10,000s of generations. We evaluate their circuit

complexity, using four information-theoretical measures, including one that emphasizes the extent to which

any network is an irreducible entity. We find that their minimal complexity increases with their fitness. (ref:

http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1003111)

iPad App development

March 2013 to Present

Page2

Members:Nikhil Joshi, Patrick Schweizer, Michael Hill

Since March 2013, I am working as the iPad app developer for the start-up labmil, LLC. I have been recently

offered partnership as the CTO for the same. The app is in its final testing phase and soon to be released, if

not sooner.

Coming from C++ background, when approached, I decided to try this new excitement, and a chance to learn

something new (Obj-C). It has been fun working on this project.

Due to ethical and business constraints I am unable to produce more information at this moment. But, sure

enough it will be exploded with remarks (if not user bug-reports).

CP violation in B-meson system

September 2010 to Present

Members:Nikhil Joshi, Karim Trabelsi, Prof. Tariq Aziz

Refer to https://documents.epfl.ch/users/j/jo/joshi/www/fun.html (Hay! That's a needle) for a non-brainer

introduction to what we did.

Laws of reliable information transfer in neuronal networks

April 2013 to Present

Members:Nikhil Joshi

The brain not only computes, but computes reliably. It forms memories as well as a rich internal

representation of the external world, allowing it to generate reliable responses to its dynamic environment.

Paradoxically, at the network level the individual neurons and synapses exhibit highly stochastic and hence

noisy behavior. Each of the many synapses impinging on a single neuron generates different activation

patterns each time the same stimulus is presented. How is it possible then, that the brain operates so reliably

when its components are highly unreliable?

In this study we seek to characterize the reliability of evoked activity motifs and their propagation in the

cortical microcircuit under activity and noise sources mimicking in vivo conditions. That is, we seek across

biophysical levels, from synapses and neurons to cortical circuits, mechanisms for transferring information in

a reliable manner using noisy elements. We adopt a combination of analytical and computational approaches

to study the phenomenon in the detailed model of cortical tissue developed at the Blue Brain Project.

Languages

English

Marathi

Hindi

French

Page3

Skills & Expertise

Physics

Statistics

Machine Learning

Programming

Matlab

Data Mining

Mathematica

Simulations

Data Analysis

Algorithms

Python

C++

Genetic Algorithms

Computational Neuroscience

R

C

Education

Tata Institute of Fundamental Research

Ph.D, Particle Physics, 2003 - 2010

Indian Institute of Technology, Kanpur

Master of Science (M.Sc.), Physics 2003, 2001 - 2003

Mumbai University

Bachelor of Science (B.Sc.), Physics, 1998 - 2001

Page4

Nikhil Joshi

Physicist, Data-scientist, Software developer

*.*.*****@*****.***

Contact Nikhil on LinkedIn

Page5

Contact this candidate