RT @aadityaura@twitter.com
I'm very happy to announce that our paper on the Distribution Shift on Question Answering Models in Clinical Domain was accepted in Robustness in Sequence Modeling at #NeurIPS2022
A short thread below summarizing the key contributions 🧵(1/n)
🐦🔗: https://twitter.com/aadityaura/status/1595359059808776192
RT @davidchalmers42
the video for my #NeurIPS2022 talk on "could a large language model be conscious?" is now online. i've given versions of the same material at adelaide, deepmind, NYU, and the #LearningSalon, but this is the best version. written version coming soon!
RT @zacknovack
1/n ‼️ Our spotlight (and now BEST POSTER!) work from the Higher Order Optimization workshop at #NeurIPS2022 is now on arxiv!
Paper 📖: https://arxiv.org/abs/2211.15853
w/@kaur_simran25 @__tm__157 @saurabh_garg67 @zacharylipton
The video of my #NeurIPS2022 talk on “building sustainable communities” is online now. I talk about two communities I help to build about 20 and 15 yrs respectively. Those are still running!
I argue for sustainable efforts, no matter how small scale they are, because over time they really deliver.
Video: https://slideslive.com/38995247
Thanks to Sara Hooker, Rosanne Liu and coorganizer for inviting me.
The video of my #NeurIPS2022 keynote on "Interaction-Centric AI" is finally publicly available. Happy to discuss & collaborate with people who find it relevant and interesting. A big thank you to Alice Oh for moderating the session.
You can now watch the recorded material from #NeurIPS2022 online without registration at:
RT @reshmigh@twitter.com
Our new work on segmenting conversational data is out on @NeurIPSConf@twitter.com
Topic Segmentation in the Wild:Towards Segmentation of Semi-structured & Unstructured Chats https://arxiv.org/abs/2211.14954
w. @BasuSamyadeep@twitter.com, Harjeet, Sharanya, Dhuri, Soundar, @UMass_NLP@twitter.com @Microsoft@twitter.com #NeurIPS2022
BYOL-Explore getting the apple! https://www.deepmind.com/publications/byol-explore-exploration-by-bootstrapped-prediction w/ Zhaohan Daniel Guo, Shantanu Thakoor, Miruna Pislar, Corentin Tallec, Florent Altché, Bernardo Avila Pires, Robin (Yunhao) Tang, Alaa Saade, Jean-Bastien Grill, Mohammad Gheshlaghi Azar, Bilal Piot, Remi Munos, Daniele Calandriello #neurips2022 #worldmodels #reinforcementlearning #ai
#NeurIPS2022 #worldmodels #reinforcementlearning #ai
For the last highlight of 2022, our latest contribution on exploiting high-level structure in #Speech for #SelfSupervised learning. Presented at #NeurIPS2022
Joint work with @tiagoCuervoG@twitter.com, Adrian Łancucki, Paweł Rychlikowski and Jan Chorowski.
Check out the nice summary 👇
https://twitter.com/tiagoCuervoG/status/1608119507519959040?t=nnqXLZCej5KRUIpQGEUYLA&s=19
#speech #selfsupervised #NeurIPS2022
@hpcprogrammer @powersoffour @jedbrown aha I tooted about my favorite #NeurIPS2022 invited talk, by Isabelle Guyon, that touched on exactly that issue: https://sites.google.com/chalearn.org/mldb/home?pli=1
"thrilled our preliminary research on #generativeai in #vr was accepted into #neurips2022
here is a bare-bones "holodeck" #3d implementation real-time in #unity. spoken user commands are compiled at runtime. 3d objects are created, loaded and manipulated via @openai's gpt-3."
Via: https://twitter.com/jasminezroberts/status/1583143311732142082
#generativeAI #vr #NeurIPS2022 #3d #unity
Watch our own @sethaxen summarize our recent #NeurIPS2022 workshop paper on modeling European #paleoclimate using #GaussianProcesses!
#GaussianProcesses #paleoclimate #NeurIPS2022
RT by @ylecun: We spoke with @alfcnz #NeurIPS2022 about his work in deep learning, his thoughts on the importance of accessibility and the use of visual formalisms to communicate complex technical information. He's also writing a book on energy based models! CC @ylecun https://www.youtube.com/watch?v=mT1Zyha0KeQ
Yann LeCun / @ylecun@twitter.com: https://twitter.com/MLStreetTalk/status/1604187472497643521#m
RT @tsitsulin_: I made an accessibility map of #Neurips2022. The country is marked in red if its resident can not get appointment in time f…
Interested in neural networks, survival analysis, and quantile regression? This paper is for you. 😁 As presented at #NeurIPS2022.
Censored Quantile Regression Neural Networks for Distribution-Free Survival Analysis
Paper: https://arxiv.org/abs/2205.13496
#MineDojo https://minedojo.org/ associates an #LLM to a #Minecraft game. Given a prompt describing an action, it will drive the character to perform them. That does not seem much, but that means the #AI is embodied in a character, which is key to grounding the knowledge.
It is trained on YouTube videos where people explain what they do.
They received #NeurIPS2022 Outstanding Paper Award.
#NeurIPS2022 #ai #minecraft #llm #minedojo
RT @MLStreetTalk
In this special edition show - we spoke with @davidchalmers42 at #NeurIPS2022 about Consciousness in Large Language Models. In this show we also discuss Nagel's bat, the Hard Problem, Zombies, and much more! with @ecsquendor and @DoctorDuggar https://youtu.be/T7aIxncLuWk
We had a great #NeurIPS2022 workshop ML for Autonomous Driving!
Most videos now here: https://ml4ad.github.io/ (slideslive videos out later)
Thank you speakers, reviewers, authors, attendees, organizers: Jiachen Li (lead), Nigamaa Nayakanti, Xinshuo Weng, Daniel Omeiza, Ali Baheri, Germán Ros
RT @MIT_CSAIL@twitter.com
An all-too-familiar problem in ML: picking hyperparameters for neural network optimizers.
MIT researchers asked, what if an optimizer could tweak its *own* hyperparameters?
Read here: Outstanding Paper Award at #NeurIPS2022 - https://bit.ly/3VTMVZx
🐦🔗: https://twitter.com/MIT_CSAIL/status/1603797104765243394
Research by H2O.ai & team worth reading is "A Brief Overview of AI governance of Responsible Machine Learning Systems"
The paper presents a framework of AI governance & its responsible use to prevent & mitigate risk.