Selected work in reinforcement learning, computer vision, and multimodal systems.
AuroraCap: Efficient, Performant Video Detailed Captioning and a New Benchmark
ICLR ’25DECORE: Deep Compression with Reinforcement Learning
CVPR ’22BDD100k: A Diverse Driving Dataset for Heterogeneous Multitask Learning
CVPR ’20Scaling Map-Elites to Deep Neuroevolution
GECCO ’20An Atari Model Zoo for Analyzing, Visualizing, and Comparing Deep Reinforcement Learning Agents
IJCAI ’19Improving Exploration in Evolution Strategies for Deep RL via a Population of Novelty-Seeking Agents
NeurIPS ’18Deep Neuroevolution: Genetic Algorithms are a Competitive Alternative for Training Deep Reinforcement Learning Agents
Best Practices for Fine-Tuning Visual Classifiers to New Domains
ECCV ’16
Vashisht MadhavanAI researcher & builder · New York