S. Chandra Mouli

AI Researcher

West Lafayette, Indiana, United States5 yrs experience

AI ML PractitionerHighly Stable

Key Highlights

Expert in machine learning and deep learning techniques.
Developed innovative forecasting methods for precision agriculture.
Contributed to Adobe's mobile application development.

Stackforce AI infers this person is a Machine Learning Specialist with experience in Agritech and Software Development.

Contact

Skills

Core Skills

Machine LearningDeep LearningAndroid Development

Other Skills

AlgorithmsArtificial Intelligence (AI)BashBlenderCC++CSSCausal InferenceCloud ComputingCollaborationCollaborative FilteringConvolutional Neural NetworksData AnalysisData MiningData Science

About

I am a Computer Science PhD candidate at Purdue University where I primarily work with Prof. Bruno Ribeiro. My current interests lie in incorporating domain knowledge in deep neural networks, for example via group-invariances or physics models, particularly for improving the extrapolation capabilities of these networks.

Experience

5 yrs

Total Experience

2 yrs 6 mos

Average Tenure

Current Experience

Purdue university

Graduate Research Assistant - SMART Films Consortium & WHIN

Aug 2018 – Jul 2022 · 3 yrs 11 mos · West Lafayette, Indiana, United States

Novel machine learning methods for accelerated testing of sensors used to measure soil properties for precision agriculture. Designed a novel physics-informed deep learning method for forecasting future sensor behavior while being robust to out-of-distribution shifts at inference time due to different environmental and manufacturing conditions. To incorporate surface roughness of the sensor for the forecasting task, used convolutional neural network over sensor images with invariance to image rotations and color shifts. Gave several poster presentations and talks at industry meetups.

Machine LearningDeep LearningPhysics-informed Deep LearningConvolutional Neural NetworksData Analysis

Adobe

Member Of Technical Staff

Jul 2015 – Aug 2016 · 1 yr 1 mo · Bengaluru Area, India

Part of a four member team working on Adobe Illustrator Draw application for Android OS. Example tasks include handling multiple drawing layers, rendering procedures for perspective/square grids for artist guidance, quick data sharing with other Adobe desktop applications, publishing the finished work to Bēhance, etc.

Android DevelopmentRendering ProceduresData SharingCollaboration

Ibm

Research Intern - IBM Research Labs

May 2013 – Jul 2013 · 2 mos · Bengaluru Area, India

Used 30 different known time series to forecast the occurrence of El Niño , a recurring climate pattern associated with warming of sea surface temperatures in the Pacific ocean, that affects climate around the world. Used Granger causality to build a graph describing the temporal relationships between the known time series and the predictand. Modeled the task as learning over a Markov random field and used Gibbs sampling for inference.
As a step toward automated machine learning (AutoML), used collaborative filtering along with meta-features of different datasets to predict the best performing machine learning algorithm for each dataset.

Time Series AnalysisGranger CausalityMarkov Random FieldsCollaborative FilteringMachine Learning