About

I'm currently a Pre-Doctoral Researcher at the Vision and AI Lab (VAL), Indian Institute of Science working with and his Ph.D. students and . I am also collaborating with from the Visual Computing Lab (VCL), Indian Institute of Science. I am supported by the Kotak IISc AI-ML Pre-Doctoral Fellowship. Before joining VAL, I worked as a research intern at Intel, Bangalore where my work focused on Continual Learning. I completed my B.Tech in Computer Science and Engineering (2018-2022) from , Bangalore where I was advised by for my Capstone Project.

CV Google Scholar ashish.ramayee at gmail dot com

Research Interests

  • Out-of-Distribution Generalization
  • Domain Adaptation
  • Foundation Models and Multimodal Learning
  • Long-tail Learning
  • Federated Learning
  • Interpretable AI

Past Experiences

News

Feb. 27, 2024 Two papers accepted at ! The first on Generalization with VLMs and the second on Long-tail learning with ViTs!
Feb. 9, 2024 Serving as a reviewer for .
Nov. 20, 2023 Presented DSiT (ICCV 2023) at the Adobe Research Workshop at IISc.
Nov. 1, 2023 Awarded the Kotak IISc AI-ML Pre-Doctoral Fellowship!
Oct. 31, 2023 Serving as a reviewer for .
Oct. 28, 2023 Our paper on Source-Free Domain Adaptation is accepted at !
Sept. 23, 2023 Serving as a reviewer for .
Jul. 18, 2023 Our paper on Source-Free Domain Adaptation is accepted at !
May 4, 2023 Serving on the Program Committee for .
Apr. 2, 2023 Our paper on Domain Generalization is accepted at the ODRUM Workshop at !
Sept. 7, 2022 Graduated from with B.Tech in CSE.
Jul. 17, 2022 Presented our paper on Interpretable AI at .
May. 16, 2022 Joined the as a Project Assistant.
Apr. 26, 2022 Our paper on Interpretable AI is accepted at .

Recent Publications

* indicates equal contribution.

More details coming soon!
Leveraging Vision-Language Models for Improving Domain Generalization in Image Classification
Sravanti Addepalli*, *, Lakshay Sharma, R. Venkatesh Babu
More details coming soon!
DeiT-LT: Distillation stikes back for Vision Transformer training on Long-tailed datasets
Harsh Rangwani, Pradipto Mondal, Mayank Mishra, , R. Venkatesh Babu

Domain-Specificity-inducing Transformers for Source-Free Domain Adaptation
Sunandini Sanyal*, *, Suvaansh Bhambri*, Akshay Kulkarni, Jogendra Nath Kundu, R. Venkatesh Babu

Aligning Non-Causal Factors for Transformer-based Source-Free Domain Adaptation
Sunandini Sanyal*, *, Suvaansh Bhambri, Pradyumna YM, Akshay Kulkarni, Jogendra Nath Kundu, R. Venkatesh Babu

Distilling from Vision-Language Models for Improved OOD Generalization in Vision Tasks
Sravanti Addepalli*, *, Lakshay Sharma, R. Venkatesh Babu
ODRUM Workshop,

Interpretability for Multimodal Emotion Recognition using Concept Activation Vectors
, K. Nidarshan, Anirudh V Ragam, Shylaja S S,

Side Work

Course and Personal Projects


Fourier Feature Mapping Networks
My implementation of the paper "Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains" by Tancik et al using Tensorflow 2.0

Semantic Segmentation
My implementation of the paper "ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation" by Paszke et al using Tensorflow 2.0

Big Data
My assignments and final project for a course on Big Data at . My final project involved Premier League Analytics using Streaming Spark.

Gait Recognition
Gait Recognition for infrared videos using Point Light Animation implemented using OpenCV. This project was carried out at the Center for Data Science and Applied Machine Learning (CDSAML) at .

This Website
A simple, responsive webpage for academics. Thanks to for this template based on .

Services

Conference Reviewer

, ,

Sub Reviewer

(Assisted my colleagues / advisor in reviewing)
, , ,

Program Committee