🗣️ Lora Aroyo – Keynote #NeurIPS2023: “The Many Faces of Responsible AI”
- callling for recognizing the inherent ambiguity of content and the natural diversity of human perspectives, urging to "treat label disagreement as signal, not noise".
neurips.cc/virtual/2023...
- callling for recognizing the inherent ambiguity of content and the natural diversity of human perspectives, urging to "treat label disagreement as signal, not noise".
neurips.cc/virtual/2023...
NeurIPS Invited Talk The Many Faces of Responsible AINeurIPS 2023
neurips.cc
July 31, 2025 at 2:50 PM
🗣️ Lora Aroyo – Keynote #NeurIPS2023: “The Many Faces of Responsible AI”
- callling for recognizing the inherent ambiguity of content and the natural diversity of human perspectives, urging to "treat label disagreement as signal, not noise".
neurips.cc/virtual/2023...
- callling for recognizing the inherent ambiguity of content and the natural diversity of human perspectives, urging to "treat label disagreement as signal, not noise".
neurips.cc/virtual/2023...
If you are interested in large language models see my paper below on how we can uncover the biases learned by these models.
Link: neurips2023-enlsp.github.io/papers/paper...
#ReinforcementLearning #FoundationModels #DeepRL #DeepReinforcementLearning #ResponsibleAI #AIBias #LLMs #LanguageModels
Link: neurips2023-enlsp.github.io/papers/paper...
#ReinforcementLearning #FoundationModels #DeepRL #DeepReinforcementLearning #ResponsibleAI #AIBias #LLMs #LanguageModels
February 11, 2025 at 5:56 PM
If you are interested in large language models see my paper below on how we can uncover the biases learned by these models.
Link: neurips2023-enlsp.github.io/papers/paper...
#ReinforcementLearning #FoundationModels #DeepRL #DeepReinforcementLearning #ResponsibleAI #AIBias #LLMs #LanguageModels
Link: neurips2023-enlsp.github.io/papers/paper...
#ReinforcementLearning #FoundationModels #DeepRL #DeepReinforcementLearning #ResponsibleAI #AIBias #LLMs #LanguageModels
Publications:
[P1] NeurIPS 2022
arxiv.org/abs/2205.14027
[P2] NeurIPS2023
arxiv.org/abs/2302.02004
[P3] ICML2024
arxiv.org/abs/2312.13426
[P4] NeurIPS2023
arxiv.org/abs/2306.04520
[P5] ICLR 2024
arxiv.org/abs/2307.09912
[P6] NeurIPS2024
arxiv.org/abs/2405.12940
[P1] NeurIPS 2022
arxiv.org/abs/2205.14027
[P2] NeurIPS2023
arxiv.org/abs/2302.02004
[P3] ICML2024
arxiv.org/abs/2312.13426
[P4] NeurIPS2023
arxiv.org/abs/2306.04520
[P5] ICLR 2024
arxiv.org/abs/2307.09912
[P6] NeurIPS2024
arxiv.org/abs/2405.12940
Learning Dynamical Systems via Koopman Operator Regression in Reproducing Kernel Hilbert Spaces
We study a class of dynamical systems modelled as Markov chains that admit an invariant distribution via the corresponding transfer, or Koopman, operator. While data-driven algorithms to reconstruct s...
arxiv.org
January 15, 2025 at 2:34 PM
Publications:
[P1] NeurIPS 2022
arxiv.org/abs/2205.14027
[P2] NeurIPS2023
arxiv.org/abs/2302.02004
[P3] ICML2024
arxiv.org/abs/2312.13426
[P4] NeurIPS2023
arxiv.org/abs/2306.04520
[P5] ICLR 2024
arxiv.org/abs/2307.09912
[P6] NeurIPS2024
arxiv.org/abs/2405.12940
[P1] NeurIPS 2022
arxiv.org/abs/2205.14027
[P2] NeurIPS2023
arxiv.org/abs/2302.02004
[P3] ICML2024
arxiv.org/abs/2312.13426
[P4] NeurIPS2023
arxiv.org/abs/2306.04520
[P5] ICLR 2024
arxiv.org/abs/2307.09912
[P6] NeurIPS2024
arxiv.org/abs/2405.12940
I don't think I'll make it to any more conferences for a while, so I'll just drop this link from our workshop at NeurIPS2023 where I presented some of these ideas for the first time.
neurips.cc/virtual/2023...
It's an exciting time to do science.
#LLMs #AI #DeepLearning #NeurIPS
neurips.cc/virtual/2023...
It's an exciting time to do science.
#LLMs #AI #DeepLearning #NeurIPS
Heavy Tails in ML: Structure, Stability, DynamicsNeurIPS 2023
neurips.cc
November 23, 2024 at 4:03 AM
I don't think I'll make it to any more conferences for a while, so I'll just drop this link from our workshop at NeurIPS2023 where I presented some of these ideas for the first time.
neurips.cc/virtual/2023...
It's an exciting time to do science.
#LLMs #AI #DeepLearning #NeurIPS
neurips.cc/virtual/2023...
It's an exciting time to do science.
#LLMs #AI #DeepLearning #NeurIPS
[2/2] In our #NeurIPS2023 paper, we introduce a simple and efficient approach to investigate how class prototype representations emerge in vision transformers trained for image classification.
arxiv.org/abs/2310.18969
arxiv.org/abs/2310.18969
Analyzing Vision Transformers for Image Classification in Class Embedding Space
Despite the growing use of transformer models in computer vision, a mechanistic understanding of these networks is still needed. This work introduces a method to reverse-engineer Vision Transformers t...
arxiv.org
November 17, 2024 at 2:06 PM
[2/2] In our #NeurIPS2023 paper, we introduce a simple and efficient approach to investigate how class prototype representations emerge in vision transformers trained for image classification.
arxiv.org/abs/2310.18969
arxiv.org/abs/2310.18969
I tried hard on the other side to figure out why the claims don’t make any sense to me and why the paper seems to misrepresent Cantar et al., NeurIPS2023, but no success.
October 22, 2024 at 12:49 AM
I tried hard on the other side to figure out why the claims don’t make any sense to me and why the paper seems to misrepresent Cantar et al., NeurIPS2023, but no success.
最新の自動運転研究 in NeurIPS2023 / zenn.dev/turing_motor...
December 28, 2023 at 3:44 AM
最新の自動運転研究 in NeurIPS2023 / zenn.dev/turing_motor...
NeurIPS2023は3,540accepted papersらしいので、そろそろ誰か目に付く論文タイトルの生成モデルをですね
December 23, 2023 at 8:07 AM
NeurIPS2023は3,540accepted papersらしいので、そろそろ誰か目に付く論文タイトルの生成モデルをですね
🎉 @SungdukYu @SciPritchard + a huge team of interdisciplinary researchers won top prize in the dataset category @ #NeurIPS2023 last week for their publication, "ClimSim: A large multi-scale dataset for hybrid physics-ML climate emulation."
👀 Read more @
👀 Read more @
November 19, 2024 at 10:07 PM
🎉 @SungdukYu @SciPritchard + a huge team of interdisciplinary researchers won top prize in the dataset category @ #NeurIPS2023 last week for their publication, "ClimSim: A large multi-scale dataset for hybrid physics-ML climate emulation."
👀 Read more @
👀 Read more @
I had **the best time** at #NeurIPS2023--more fun than I've ever had--and bummed it's over. If you're a tech minority, maybe you can come next year? It will be in Vancouver, Canada. Below, I'll detail themes where it seems the majority of research comes from tech minorities: 🧵
December 18, 2023 at 12:03 AM
I had **the best time** at #NeurIPS2023--more fun than I've ever had--and bummed it's over. If you're a tech minority, maybe you can come next year? It will be in Vancouver, Canada. Below, I'll detail themes where it seems the majority of research comes from tech minorities: 🧵
Final day in New Orleans — what an amazing city! #neurips2023
December 17, 2023 at 2:26 PM
Final day in New Orleans — what an amazing city! #neurips2023
Farewell #NeurIPS2023 and NOLA. It's been a blast: meeting so many old friends and making new ones. So many interesting workshops, it is the best part of the meeting
Have A Wondeful Trip And Safe Travels GIF
Posted from Graysky - get the app to view and post GIFs!
graysky.app
December 16, 2023 at 10:36 PM
Farewell #NeurIPS2023 and NOLA. It's been a blast: meeting so many old friends and making new ones. So many interesting workshops, it is the best part of the meeting
Join our poster on "Data Augmentations in Deep Weight Spaces" @neur_reps workshop, #NeurIPS2023, Ballroom A+B at 4pm!
February 7, 2025 at 10:19 AM
Join our poster on "Data Augmentations in Deep Weight Spaces" @neur_reps workshop, #NeurIPS2023, Ballroom A+B at 4pm!
Great collaboration and promising results on the benefits of recursive ridge regression for tackling data heterogeneity in federated learning.
Led by @ErosFani and presented today by @marcuswallacej at the Workshop on FL in the Age of Foundation Models @ #NeurIPS2023!
x.com/mciccone_AI/st…
Led by @ErosFani and presented today by @marcuswallacej at the Workshop on FL in the Age of Foundation Models @ #NeurIPS2023!
x.com/mciccone_AI/st…
December 23, 2024 at 10:50 AM
Great collaboration and promising results on the benefits of recursive ridge regression for tackling data heterogeneity in federated learning.
Led by @ErosFani and presented today by @marcuswallacej at the Workshop on FL in the Age of Foundation Models @ #NeurIPS2023!
x.com/mciccone_AI/st…
Led by @ErosFani and presented today by @marcuswallacej at the Workshop on FL in the Age of Foundation Models @ #NeurIPS2023!
x.com/mciccone_AI/st…
NeurIPS 2023 Day 7 (Main Conference + Workshops) in New Orleans, LA still going strong. Even on the last day, sessions are well-attended. #NeurIPS2023
December 16, 2023 at 3:26 PM
NeurIPS 2023 Day 7 (Main Conference + Workshops) in New Orleans, LA still going strong. Even on the last day, sessions are well-attended. #NeurIPS2023
Last night in New Orleans for #NeurIPS2023 A legendary place with amazing southern food and drinks! Cheers to successful conference and Armed Forces of Ukraine for ensuring safety of my family and friends 💪🇺🇦
December 16, 2023 at 3:46 AM
Last night in New Orleans for #NeurIPS2023 A legendary place with amazing southern food and drinks! Cheers to successful conference and Armed Forces of Ukraine for ensuring safety of my family and friends 💪🇺🇦
After #NeurIPS2023 I am very tempted to write a paper titled “Quantum Mechanics a All You Need” 🧐#compchem #chemsky
December 16, 2023 at 1:28 AM
After #NeurIPS2023 I am very tempted to write a paper titled “Quantum Mechanics a All You Need” 🧐#compchem #chemsky
The new Moore's Law of generative AI: the number of generative and docking models is doubling every #NeurIPS2023 #NeurIPS23
December 15, 2023 at 11:17 PM
The new Moore's Law of generative AI: the number of generative and docking models is doubling every #NeurIPS2023 #NeurIPS23
Great talk by Dr. Sebastian Raschka at their #NeurIPS2023 competition workshop in New Orleans, LA on using LoRA in finetuning LLMs. He shared a long list of results of experiments on parameter settings. LoRA was used by the competition winner. #LLMs #largelanguagemodel
December 15, 2023 at 8:42 PM
Great talk by Dr. Sebastian Raschka at their #NeurIPS2023 competition workshop in New Orleans, LA on using LoRA in finetuning LLMs. He shared a long list of results of experiments on parameter settings. LoRA was used by the competition winner. #LLMs #largelanguagemodel
Today, the workshops started at #NeurIPS in New Orleans, Louisiana. I chose to attend the Generative AI in Education. #ChatGPT has already invaded our classes. Dr. Elena Glassman (Harvard) details research findings on how to uphold learning in today’s environment. #NeurIPS2023
December 15, 2023 at 3:57 PM
Today, the workshops started at #NeurIPS in New Orleans, Louisiana. I chose to attend the Generative AI in Education. #ChatGPT has already invaded our classes. Dr. Elena Glassman (Harvard) details research findings on how to uphold learning in today’s environment. #NeurIPS2023
-Tomorrow- at #NeurIPS2023 you can catch me in Hall B2 at the Robot Learning Workshop bright and early at 8:45am-9:10am talking about how "LPTMs Can Help Robots Without Ignoring Robotics", and where I feel the community (and my own lab!) have mostly missed the mark here so far.
December 15, 2023 at 3:44 PM
-Tomorrow- at #NeurIPS2023 you can catch me in Hall B2 at the Robot Learning Workshop bright and early at 8:45am-9:10am talking about how "LPTMs Can Help Robots Without Ignoring Robotics", and where I feel the community (and my own lab!) have mostly missed the mark here so far.
Cutting- edge research
TIES-Merging removes model merging interference #NeurIPS2023 👇
arxiv.org/abs/2306.01708
Thanks for the pic
TIES-Merging removes model merging interference #NeurIPS2023 👇
arxiv.org/abs/2306.01708
Thanks for the pic
December 15, 2023 at 2:51 PM
Cutting- edge research
TIES-Merging removes model merging interference #NeurIPS2023 👇
arxiv.org/abs/2306.01708
Thanks for the pic
TIES-Merging removes model merging interference #NeurIPS2023 👇
arxiv.org/abs/2306.01708
Thanks for the pic
Models can't decide what to do with the update?
Let them vote.
Delete small (noise) parameters
Delete weights with sign disagreements
Sounds interesting?
What if we say model merging\fusion?
arxiv.org/abs/2306.01708
Come talk in the poster
#NeurIPS2023
Poster #1118 enjoy the walk...
Let them vote.
Delete small (noise) parameters
Delete weights with sign disagreements
Sounds interesting?
What if we say model merging\fusion?
arxiv.org/abs/2306.01708
Come talk in the poster
#NeurIPS2023
Poster #1118 enjoy the walk...
TIES-Merging: Resolving Interference When Merging Models
Transfer learning - i.e., further fine-tuning a pre-trained model on a downstream task - can confer significant advantages, including improved downstream performance, faster convergence, and...
arxiv.org
December 14, 2023 at 9:36 PM
Models can't decide what to do with the update?
Let them vote.
Delete small (noise) parameters
Delete weights with sign disagreements
Sounds interesting?
What if we say model merging\fusion?
arxiv.org/abs/2306.01708
Come talk in the poster
#NeurIPS2023
Poster #1118 enjoy the walk...
Let them vote.
Delete small (noise) parameters
Delete weights with sign disagreements
Sounds interesting?
What if we say model merging\fusion?
arxiv.org/abs/2306.01708
Come talk in the poster
#NeurIPS2023
Poster #1118 enjoy the walk...
Everybody meet my new pet!
We met in the exhibit hall at Disney Research @NeurIPS2023
We met in the exhibit hall at Disney Research @NeurIPS2023
December 14, 2023 at 9:24 PM
Everybody meet my new pet!
We met in the exhibit hall at Disney Research @NeurIPS2023
We met in the exhibit hall at Disney Research @NeurIPS2023