- Deep Learning for Graphs (Geometric Deep Learning)
- Social Graph Analysis
- Data Mining
- Visual Question Answering
- Computer Vision
- Indian Institute of Technology Hyderabad
- M.Tech (RA) in Computer Science and Engineering
- Aug, 2015 - Present
- CGPA: 9.29 / 10
- Vishwakarma Institute of Technology Pune
- B.Tech. in Computer Science
- Aug, 2010 - May, 2014
- CGPA: 9.02 / 10
- Research Assistant (Aug, 2015 - Present)
- Organization: Indian Institute of Technollogy Hyderabad
- Supervisor: Vineeth N Balasubramanian.
- Research Intern (May, 2017 - July, 2017)
- Data Analyst Intern (Dec, 2016)
- Organization: Suzuki, Hamamatsu, Japan
- Analyzed the data from different social media: Twitter, YouTube, Blog Posts, Online product review sites.
- Used Radian6 and QlikView for data analysis and visualization.
- Supervisor: Shoji Kato-san
- Summer Intern (June, 2016 - July, 2016)
- Organization: Bosch, Bangalore
- Designed Machine Learning algorithms for analysis of sensor data using Python-ScikitLearn, Pandas and Unscrambler.
- Supervisor: Tony Francis
- Software Developer (Aug, 2014 - July, 2015)
- Organization: Persistent Systems, Pune
- Worked as Java Developer
- Supervisor: Taraben Khoiwala
Pandhre, Supriya, et al. “STwalk: learning trajectory representations in temporal graphs.” Proceedings of the ACM India Joint International Conference on Data Science and Management of Data. ACM, 2018.
Pandhre, Supriya, and Shagun Sodhani. “Survey of Recent Advances in Visual Question Answering.” arXiv preprint arXiv:1709.08203 (2017).
Pandhre, Supriya, Manish Gupta, and Vineeth N. Balasubramanian. “Community-based Outlier Detection for Edge-attributed Graphs.” arXiv preprint arXiv:1612.09435 (2016).
- Using Generative Adversarial Network (GAN) for Image Completion:
- Completed as a course project for Deep Learning for Vision course
- Goal is to use GANs for completion of face images, important in many applications such as surveillance.
- Implemented in Python-TensorFlow.
- Mentor: Vineeth N Balasubramanian.
- Inverse Search:
- Completed as a course project for Information Retrieval course
- The goal is, given a set of documents, find the most probable query that might have generated these documents.
- Used ensemble of 3 algorithms:TF-IDF based, LDA based and TextRank.
- Implemented using Python NLTK, Java MALLET.
- Mentor: Maunendra Desarkar
- Deep Learning Libraries:
- Data Analysis and Visualization tools:
- Python, ScikitLearn
- Radian 6
- Java EE