CV

Areas of Interest

Computer Vision, Deep Learning, Generative AI, Machine Learning, Medical Imagery, AI for Social Good, Trustworthy AI, Robust Neural Networks, Natural Language Processing, Graph Neural Networks, Data Analysis.

Publications

Evaluating Predictive Uncertainty and Robustness to Distributional Shift Using Real World Data

Performance Evaluation of Deep Segmentation Models on Landsat-8 Imagery

Akshat Bhandari(∗), Sriya Rallabandi(∗), Sanchit Singhal(∗), Aditya Kasliwal(∗), and Pratinav Seth. Performance evaluation of deep segmentation models on landsat-8 imagery. (* - equal contribution) Tackling Climate Change with Machine Learning Workshop, NeurIPS 2022, 2022. URL: https://www.climatechange.ai/papers/neurips2022/92

Uncertainty Aware Test Time Augmented Ensemble for PIRC Diabetic Retinopathy Detection

Pratinav Seth, Adil Hamid Khan(∗), Ananya Gupta(∗), Saurabh Mishra(∗), and Akshat Bhandhari. Uatta-ens: Uncertainty aware test time augmented ensemble for pirc diabetic retinopathy detection. (* - equal contribution) Medical Imagery Meets NeurIPS Workshop,NeurIPS 2022, 2022. URL: http://www.cse.cuhk.edu.hk/∼qdou/public/medneurips2022/95.pdf

Uncertainty-aware test-time augmented ensemble of berts for classification of common mental illnesses on social media posts

Pratinav Seth(∗) and Mihir Agarwal(∗) . Uncertainty-aware test-time augmented ensemble of berts for classification of common mental illnesses on social media posts. In Krystal Maughan, Rosanne Liu, and Thomas F. Burns, editors, (* - equal contribution) The First Tiny Papers Track at ICLR 2023, Tiny Papers @ ICLR 2023, Kigali, Rwanda, May 5, 2023, 2023. URL: https://openreview.net/pdf?id=a9VgV-hywP

Corefusion: Contrastive regularized fusion for guided thermal super-resolution

Aditya Kasliwal, Pratinav Seth, Sriya Rallabandi, and Sanchit Singhal. Corefusion: Contrastive regularized fusion for guided thermal super-resolution. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshop (CVPRW), pages 507–514, 2023. URL: https://ieeexplore.ieee.org/document/10208919

SSS at SemEval-2023 task 10: Explainable detection of online sexism using majority voted fine-tuned transformers

Sriya Rallabandi, Sanchit Singhal, and Pratinav Seth. SSS at SemEval-2023 task 10: Explainable detection of online sexism using majority voted fine-tuned transformers. In Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023), pages 1231–1236, Toronto, Canada, July 2023. Association for Computational Linguistics. URL: https://aclanthology.org/2023.semeval-1.171, doi:10.18653/v1/2023.semeval-1.171

Refuseg: Regularized multi-modal fusion for precise brain tumour segmentation

Aditya Kasliwal, Sankarshanaa Sagaram, Laven Srivastava, Pratinav Seth, and Adil Khan. Refuseg: Regularized multi-modal fusion for precise brain tumour segmentation. Accepted at 9th Brain Lesion (BrainLes) workshop - the satellite event of the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2023. URL: https://aps.arxiv.org/pdf/2308.13883.pdf

RSM-NLP at BLP-2023 Task 2: Bangla Sentiment Analysis using Weighted and Majority Voted Fine-Tuned Transformers

Pratinav Seth, Rashi Goel, Komal Mathur and Swetha Vemulapalli. RSM-NLP at BLP-2023 Task 2: Bangla Sentiment Analysis using Weighted and Majority Voted Fine-Tuned Transformers. Conditionally Accepted at Proceedings of the 1st Workshop on Bangla Language Processing (BLP 2023), EMNLP, Association for Computational Linguistics.

Academic Awards

• Accepted as an AAAI Undergraduate Consortium Scholar to present my Work on Uncertainty Quantification. Includes scholarship to attend AAAI-23 (Feb. 2023)
• Received an Undergraduate Research Grant worth 10000 INR from MAHE to work on Explainable & Trustworthy Skin Lesion Classification (Jan. 2023)
• Among the 35 Students(incld. MS, PhDs) invited to Attend MSR India & UPenn Workshop on Trustworthy AI (Jan. 2023)
• Placed 5th among 134 Teams, 13th International Cyber Security Data Mining Competition 2022 (Nov. 2022)

Work experience

• Computer Vision Research Engineer Intern (CR/RDT-2), Bosch Corporate Research (Robert Bosch Research and Technology Center India) Bangalore, India (June 2023 – October 2023):
  • Worked on Generative Modelling Applications using GANs, Latent Diffusion, and Stable Diffusion Models with applications on Domain Adaptation and Robust Computer Vision pipelines in Automobile Urban Scene Datasets.
  • Supervisor: Koustav Mullick (CR/RDT-2), Computer Vision Researcher & Activity Manager, Bosch Corporate Research India.
• Machine Learning Intern, Eedge.ai (March 2022 - May 2022):
  • Worked on the product development of deep learning-backed intelligent platform .
  • Used Data Analysis, Machine Learning, Generative Models, Explainable AI, and Deep Learning Techniques to improve performance.
• Data Science Intern, CUREYA(Aspexx Health Solutions Private Limited) (Jan. 2022 - Feb. 2022):
  • Worked on the product development of in-house conversational Healthcare chatbot Reyana with custom text understanding and response generation based on the type of user response.
  • Implemented Text Classification Algorithms, Text Summarizing Algorithms, BERT Models, and LSTM-based NLG techniques

Research Experience

• Research Intern, KLIV Research Group, Indian Institutes of Technology Kharagpur (May 2022 - Sep. 2023):
  • Working on applications of Trustworthy and explainable Machine Learning methods in Medical Imagery involving Chest Radiographs with a focus on the applications of multi-class noisy label problems using Graph Neural Networks, Self-Supervised Learning Methodologies, and Vision Attention Models with applications in Radiographs based Machine Learning.
  • Supervisor: Rakshith Satish, MS Scholar , IIT Kharagpur & PI: Dr. Debdoot Sheet, IIT Kharagpur
• Research Collaborator, Indian Institutes of Technology Roorkee (Oct. 2022 - Mar. 2023):
  • Worked in the domain of Deep Metric Learning involving SSL, Contrastive Learning, and Vision Attention Models.
  • Supervisor: Dr. Vijay Kumar BR ,NEC Labs & Anmol Agarwal, IIT Roorkee
• Research Assistant, Manipal Institute of Technology (Dec. 2022 - Oct. 2023):
  • Worked on skin-tone aware skin lesion classification using vision attention models and ExplainableAI techniques.
  • Worked on the fairness of skin lesion classification systems using SSL,MiM & ViT to prevent the development of bias due to skin tone. This project is funded with support from MAHE UG Research Grant worth 10000 INR.
  • Supervisor: Dr. Abhilash Pai, Dept. of Data Science and Computer Application, Manipal Institute of Technology, MAHE.
• Research Assistant, Manipal Institute of Technology (June 2022 - Sep. 2022):
  • Working on an intersection of Cybersecurity and Artificial Intelligence involving Multi-class malware classification using various Deep Learning and Machine Learning techniques.
  • Worked on Multi-class malware classification. Applied various Deep Learning and Machine Learning techniques. Placed 5th among 134 Teams, 13th International Cyber Security Data Mining Competition 2022:
  • Supervisor: Dr.Vidya Rao & Dr. Poornima Panduranga Kundapur, Dept. of Data Science and Computer Application, Manipal Institute of Technology, MAHE.

Academic Volunteering & Services

• Program Committee at SyntheticData4ML Workshop , NeurIPS 2022 (Oct. 2022)
• Reviewer at Topological, Algebraic, and Geometric P.R.A. Workshop, CVPR 2023 (Apr. 2023)
• Program Committee at Domain Adaptation and Representation Transfer Workshop, MICCAI 2023 (July. 2023)
• Program Committee at SyntheticData4ML Workshop , NeurIPS 2023 (Oct. 2023)

Hackathon Awards

• Placed among Top 10 teams out of 1000 submissions in Bajaj Finserv HackRx3.0 Hackathon (June 2022)
• 3rd place in Engima Hackathon in Prometheus X by IECSE Manipal (July 2021)
• 1st Place IEEE SB Manipal Machine Learning Challenge Techno Colosseum (July 2021)

Education

• B.Tech in Data Science, Manipal Institute of Technology (2020 - 2024 {#})
  • Relevant Coursework - Deep Learning, Machine Learning, Data Analytics, Multivariate Statistics, Probability, Econometrics, Graph Theory, Linear Algebra.

Volunteering, Service and Leadership

• Co-President, The Research Society MIT (Aug. 2022 – Sep. 2023)
  • Led and managed the Research Society MIT, an undergraduate research organization with over 90 members from 10+ technical domains. Enabled multidisciplinary collaboration and communication among members of various domains. Recruited new members from a pool of 250+ applicants. Managed the learning and development of new members. Spread awareness about the importance and accessibility of undergraduate research among university students. Worked on several interdisciplinary research projects, including contributions to the Artificial Intelligence flank. Mentored 10+ sophomore and junior undergraduates in various topics of Artificial Intelligence.
• Co-founder & Head of Artificial Intelligence, The Data Alchemists (Nov. 2022 – Sep. 2023)
  • Co-founded and served as the Head of Artificial Intelligence and Machine Learning, responsible for overseeing the club’s AI and ML division. Recruited the first members of the club and helped to grow its membership to over 30 members. Took online workshops on machine learning and its applications and organized workshops and events for club members. Helped club members to work on AI projects in various aspects of machine learning, natural language processing and computer vision.
• Senior Researcher & Co-Lead of Artificial Intelligence Research Wing, Mars Rover Manipal (Aug. 2022 – Aug. 2023)
  • As a senior member of the research sub-division, I collaborated with and guided six sophomore and junior undergraduates in their research journey. I worked on projects in machine learning, deep learning, computer vision, and natural language processing, with a keen focus on medical imagery, robust neural networks, synthetic data, semantic segmentation, data fusion, and sentiment analysis. This led to 4+ workshop publications in CVPR, ACL & NeurIPS conferences during my tenure.
• Junior Researcher in Artificial Intelligence Research Wing, Mars Rover Manipal Aug. 2021 – Aug. 2022
  • Worked under the guidance of the research head on various projects in machine learning, deep learning, computer vision, and natural language processing. Had a keen focus on non-Bayesian uncertainty quantification, medical imagery, robust neural networks, generative models, meta-learning, and sentiment analysis. Led to a workshop publication at the NeurIPS conference, a pre-print, and participation in various shared tasks and competitions in AAAI and NeurIPS during my tenure. Managed a recruitment drive for the division, increasing applications from 50 to 200+ applicants. Led a training phase of 50+ candidates over six months in various aspects of machine learning and selected 6 for the next tenure.
• Student Trainee in Artificial Intelligence Research Wing, Mars Rover Manipal Dec. 2020 – Jul. 2021
  • Worked as a taskphase trainee under the guidance of junior researchers in the AI Research Wing. I learned about various topics in probability, statistics, machine learning, deep learning, natural language processing, reinforcement learning, and generative models through bi-weekly discussions and hands-on projects that implemented theoretical concepts. I also participated in weekly paper discussion meetings with electronics and mechanical domain experts to develop multidisciplinary ideas
• Head Boy, Bhavan’s G.K. Vidyamandir (Nov. 2018 – Nov. 2019)
  • Managed the every day and several special events of the school while supervising a team of more than 50 students.

Skills

• Languages:
  • Python
  • C++
  • SQL
  • Java
  • C
  • Latex
• Frameworks :
  • TensorFlow,Keras,PyTorch
  • Scikit-Learn
  • NLTK,SpaCy
  • NetworkX
  • Dask.
• Libraries:
  • Numpy,Pandas,Seaborn,Matplotlib.
  • OpenCV,PIL.
  • Huggingface.
  • SHAP,LIME.
• Tools& Misc.:
  • HTML,CSS,
  • Git,
  • Jupyter Notebook,Google Colab.
  • Linux, Windows.
  • Weka,Excel.
  • Azure.
• MOOC: Deep Learning Specialization - DeepLearning.ai
• Misc.: Attended 6th Summer School on AI organized by- CVIT IIITH