Tarun
@_taruntino
Followers
137
Following
971
Media
0
Statuses
45
Research Scientist @waymo. Past: @NVIDIAAI @MSFTResearch, PhD in ML at @UniOfOxford @ExeterCollegeOx. Clarendon scholar.
Oxford, England
Joined April 2013
Riding in a @Waymo autonomous vehicle is an amazing futuristic experience, straight out of science fiction. So excited to see the service coming to my hometown in London 🇬🇧! Can't wait for Londoners to try them out soon!
102
228
3K
Waymo is coming to London next year.
waymo.com
Waymo is expanding to London, with plans to offer rides starting in 2026
1
4
44
We’re excited to share that the 2025 Waymo Open Dataset Challenges are now open. Our 6th annual edition features four research topics: • Vision-Based End-to-End Driving • Scenario Generation • Interaction Prediction • Sim Agents Learn more: https://t.co/yRr0EVPPtb
17
23
100
A paper in @Nature presents a generative AI tool that could help video game designers to craft gameplay iteratively. The AI model generated robust 3D worlds that obeyed the mechanics of the video game they were designed for. https://t.co/UbDF4nkdN2
1
13
48
Today in Nature: our research on world and human action models (WHAM) - generative ai models of video games, aimed towards supporting game creatives in gameplay ideation : https://t.co/zeaOGbfVqh - huge congrats to everyone who made this happen, I couldn't be more proud 🥳
nature.com
Nature - A state-of-the-art generative artificial intelligence model of a video game is introduced to allow the support of human creative ideation, with the analysis of user study data highlighting...
20
56
220
Super excited to share that the paper “World and Human Action Models towards gameplay ideation” is now published in Nature!🌐🎮 Weights released on @huggingface (200M and 1.6B) Paper: https://t.co/GMr1iRdp4C Model: https://t.co/vpeTQjV2Db A massive thank you to the team!🙏🥰
huggingface.co
Nature published Microsoft research detailing our WHAM, an AI model that generates video game visuals & controller actions. We're releasing the model weights, sample data & WHAM Demonstrator on Azure AI Foundry to enable researchers to build on this work. https://t.co/R3Pt8iwPBH
0
2
16
Proud to have been part of this journey as an intern and to see it come so far! Huge shoutout to an amazing team—great job!
If you thought AI-generated text, images, and video were cool, just imagine entire interactive environments like games!
0
0
12
Nature published Microsoft research detailing our WHAM, an AI model that generates video game visuals & controller actions. We're releasing the model weights, sample data & WHAM Demonstrator on Azure AI Foundry to enable researchers to build on this work. https://t.co/R3Pt8iwPBH
1
19
79
My team at Waymo is hiring two London-based interns for Summer 2025. Please consider applying! https://t.co/OHumU8IA8B
https://t.co/KGrbCVqmuQ
1
11
63
Video models like Sora and Gen 3 can generate realistic videos, but can they produce useful synthetic data for planning/RL? Our work (AVID) explores how pretrained image-to-video models can be adapted to accurate action-conditioned world models. 1/n
7
19
106
Great opportunity with a great team!
It's that time of year again! We've just announced our game intelligence research internship - join us to learn, work with a fantastic team, and tackle hard problems. "Internship Opportunity: Research Intern – Multimodal Generative Models for Video Games"
0
0
3
In case you missed it in the ICML rush, we released a meta-RL survey paper! 🤖 Work by @jakeABeck, @ristovuorio, Zheng Xiong, @luisa_zintgraf, @shimon8282, in addition to our wonderful collaborators at Stanford, Evan Liu and @chelseabfinn Highlights in this thread ⬇️
Excited to share our new survey paper of meta-RL! 📊🤖🎊 https://t.co/R3qHbNTGnW Many thanks to my co-authors for the hard work, @ristovuorio, Evan Liu, Zheng Xiong, @luisa_zintgraf, @chelseabfinn, @shimon8282 Highlights in the thread below!
0
8
19
We are looking for candidates for our exciting new (fully paid) #PhD position on "reliable deep multi-task #reinforcementlearning learning using epistemic #uncertainty estimation in #graphneuralnetworks" at #TUDelft (Netherlands)!
2
6
12
Check out our NeurIPS paper Foundation Models for Semantic Novelty in RL! We use language abstractions from foundation models to explore! At the Workshop on Foundation Models for Decision Making https://t.co/nXkNTfP25d -@tarun_gup, Peter Karkus, Tong Che, @danfei_xu, @drmapavone
1
4
4
🧵THREAD 🧵 Are you looking to do a 4 year Industry/Academia PhD? I am looking for 1 student to pioneer our new FAIR-Oxford PhD programme, spending 50% if their time at @UniofOxford, and 50% at @facebookai (FAIR) while completing a DPhil (Oxford PhD). Interested? Read on… 1/9
9
110
460
Four impressive papers from our lab were accepted to NeurIPS this year! 🎊🎉🎊 Check out their outstanding work below 👇
1
2
29
I am hiring a postdoc in deep reinforcement learning! This is a purely academic post in my @whi_rl lab at Oxford but funded by a generous grant from @Waymo. The project will focus on multi-agent, Bayesian, and/or meta reinforcement learning. https://t.co/wx8LoZKQwY
cs.ox.ac.uk
152350
2
73
184
UneVEn executes similar solved tasks with highest Q-values which has a high chance to execute optimal actions of nearby, currently unsolvable tasks, and then use the sampled actions to learn on all tasks. This increases the set of solvable tasks until we reach target task. (5/)
0
0
2
We propose a novel exploration scheme (UneVEn) which tackles the RO problem by sampling nearby similar tasks related to the target task which might be less prone to RO. (4/)
1
0
2
Current SOTA MARL methods fail to learn on tasks requiring high degree of simultaneous coordination by agents due to relative overgeneralization (RO). This leads to suboptimal policies especially in tasks having high penalty for miscoordination. (3/)
1
0
1