Arun Ramachandran

I'm a machine learning research engineer at MIT Lincoln Laboratory. I work in the Homeland Protection Systems Group on applying ML research to problems of national security.

I graduated in May 2023 with a Master's in Computer Science from Georgia Tech. I was advised by Prof. Dhruv Batra.

I am interested in embodied AI — developing agents that can see, act and reason in the world. Specifically, I am interested in areas of embodied AI related to language, commonsense reasoning and generalization capabilities.

In a past life, I was a software engineer at Microsoft, working in the Bing Shopping team. I was initially a full-stack web developer and later moved into machine learning. I worked on applying natural language processing and information retrieval techniques to areas like query understanding and product ranking. I also did independent research on emotion analysis in low-resource languages.

I did my undergrad in Computer Science and Engineering at NIT Trichy. During that time, I interned at Amazon and Microsoft, working on Alexa and Cortana. My undergrad thesis focused on an empirical analysis of the effectiveness of different game tree search algorithms.

2014 - 2018
Summer 2016
Summer 2017,
2018 - 2021
2021 - 2023
Summer - Fall 2022
2023 - Now

Publications(*=equal contribution)

HomeRobot: Open-Vocabulary Mobile Manipulation

NeurIPS Competition Track 2023, CoRL 2023

Sriram Yenamandra*, Arun Ramachandran*, Karmesh Yadav*, Austin Wang, Mukul Khanna, Theophile Gervet, Tsung-Yen Yang, Vidhi Jain, Alexander William Clegg, John Turner, Devendra Singh Chaplot, Zsolt Kira, Manolis Savva, Angel Chang, Dhruv Batra, Roozbeh Mottaghi, Yonatan Bisk, Chris Paxton

Paper Project Page
Housekeep: Tidying Virtual Households using Commonsense Reasoning

ECCV 2022

Yash Kant, Arun Ramachandran, Sriram Yenamandra, Igor Gilitschenski, Dhruv Batra, Andrew Szot*, Harsh Agrawal*

Paper Project Page
Cross-Lingual Emotion Lexicon Induction using Representation Alignment in Low-Resource Settings


Arun Ramachandran, Gerard de Melo

Paper Project Page


Improving the Student-Teacher Approach for Semi-Supervised Semantic Segmentation

Arun Ramachandran, Jeremiah Coholich, Harsh Maheshwari, Simar Kareer

We propose and evaluate two variants to the mean teacher self-distillation technique: sliding-window teacher evaluation and the ensembling of two mean-teachers with different update rates. The latter modification improved results over the baseline mean-teacher.

fasttensor: A C++ high-performance computing library

I wrote a C++ library for high-performance computing of basic arithmetic operations on tensors. It uses SIMD instructions on CPU and CUDA on GPU to accelerate computation, and can use multiple GPUs. It uses expression templates and kernel fusion to avoid unnecessary memory allocations and copies. It achieves over 1.5x speedup over normal eager evaluation.

Chess AI: A self-learning chess artificial intelligence

I made a self-learning chess AI with reinforcement learning and game tree searching. I used MTD-f, an optimized version of minimax, to speed up game tree searching by 1.5x. The AI ended up successfully learning positional advantages of pieces in different phases of the game.

Code Character: An online AI contest

Code Character is a yearly contest hosted by my undergraduate institution, where players write their own AI to control troops in a real-time strategy game. I designed the game for the 2017 and 2018 editions and led a team of programmers to make a web app to automate submissions and a cross-platform C++ engine for the game. In 2018, we launched the contest and had over 1300 participants from 74 different countries.



Some browser games I made a long, long time ago.

Grav Guy