Me:
Currently, working at Facebook AI as a Machine Learning Researcher.
Worked at Apple between 11/2014-05/2019 as a Machine Learning Researcher.
I worked with a group of 20 researchers passionate about NLU at Nuance for about 18 months. (Mostly working on Learning under constraints using techniques from semi supervised learning, inductive logic programming, rule learning, deep-learning, sparse encoding etc. )
I finished my PHD (total 4 years in grad school) from the Department Of Computer Science at the University Of Minnesota 2013. I received my masters in computer science from this very department in 2011.
Broadly, my research interests are in the area of probabilistic AI, machine learning, natural language processing/understanding etc. More specifically I am excited about incorporating contextual information into knowledge discovery algorithms and models and hence make them human-like creative. My past works include the development of an algorithm, which finds interestingly deviant topics from text corpora, a mathematical model of humor a general theory of interesting-ness, discovery of concept focal points etc.
The list of papers and presentations have piled up heavily over the last few last few months, will update the list soon.
Here's a list of my papers:
Tech blogs
[1] Designing a constrained exploration system- Instagram's Engineering Blog
[2] On The Value Of Diversified Recommendations
[1]Journal Papers
[1.1] Amogh Mahapatra, Nisheeth Srivastava, Jaideep Srivastava. Contextual Anomaly Detection in Text Data. Algorithms Journal, Volume 5, Issue 4
[1.2] H. Chen, Y. Zhang, W. S. Bainbridge, K. J. Shim, N. Pathak, M.A. Ahmad,C. DeLong, Z. Borbora, A. Mahapatra , J. Srivastava . AI, Virtual Worlds, and Massively Multiplayer Online Games. IEEE Intelligent Systems, January/February 2011, pp. 80-89.
[2] Conference Papers
[2.1] Amogh Mahapatra, Nisheeth Srivastava, Jaideep Srivastava. Characterizing The Internet’s Sense Of Humor. IEEE SocialCom 2012 Amsterdam, NL September 3-5, 2012.
[2.2] Amogh Mahapatra, Nisheeth Srivastava, Jaideep Srivastava. Contextual Anomaly Detection in Text Data. SDM(TMW)-2012.
[2.3] Amogh Mahapatra, Xin Wan, Yonghong Tian, Jaideep Srivastava. Augmenting Image Processing with Social Tag Mining for Landmark Recognition. Proceedings of the 17th international conference on Advances in multimedia modeling,2011.
[2.4] Amogh Mahapatra, Jaideep Srivastava. Incongruity vs Incongruity Resolution. IEEE Social Computing 2013
[2.5] Mahapatra A., Kapoor K., Kasturi R., Srivastava J. Improving Classification Accuracy Of YouTube Videos by Exploiting Focal Points In Social Tags, ICME 2013
Here's a list of some of my previous projects:
1) Face Recognition using Genetic Algorithms based template matching.
2) A novel method to calculate hill-cipher secret keys using self-repetitive matrices.
3) Computing DNA copy number variation points using techniques of image processing and pattern recognition.
4) Implementing turbo codes and Viterbi algorithm in a highly noisy channel.
5) Survey of techniques used to break combinatorial nastiness in Chinese checkers.
6) Sentiment analysis from twitter feeds based on external semantic information.
7) Investigation into Multivariate Poisson distribution and Anomaly detection using it on census data.
8) Focal point learning for extracting more predictive exemplars in text data.
Industry Experience:
Previously, I worked at Microsoft India R&D for 2 years as a Software Engineer. (7/2007-7/2009)
Secondary Research Interests:
My secondary research interests are way too many, it ranges from rigorously mathematical fields like quantum computing, group theory, group character theory, cellular automaton to subtler fields like music-appreciation, Jungian psychology. In short, I enjoy inter-disciplinary research.