Gain insights on the challenges and solutions in aligning AI technologies with societal values and legal frameworks. I hope you enjoy the discussion! Full interview on YouTube: https://
youtu.be/APpB0lH8YLE
On Spotify: https://
open.spotify.com/episode/4O0dqn
CKu4URQz4qj7OHH0?si=k1uZqpKpTTuXNTRBg6LHwQ
…
RESEARCH
-
AI alignment with societal values and legal frameworks discussion
By
–
-
Ethics in AI: Deep Dive into Best Practices
By
–
Good morning fellow AI enthusiast! This week's iteration focuses on AI ethics with a new fantastic episode of the What's AI podcast with the amazing Auxane Boch! Learn more in the newsletter: https://
open.substack.com/pub/louisbouch
ard/p/ethics-in-ai-a-deep-dive-into-best?r=25qlky&utm_campaign=post&utm_medium=web
… Still at #iccv2023 if you want to hang out! -
Skin tone bias in medical AI training data impacts diagnosis
By
–
Black and brown skin tones are underrepresented in teaching materials in medicine; what does this mean for the diagnosis and treatment of patients from these communities?
-
Comparing Closed vs Open Source Models: Deployment Costs and Fairness
By
–
how can you compare a non open source, non deployable model with an open source model which is deployable? whats the actual cost fpr deploying gpt3.5 or gpt4? i also have a lot of respect for openai but not for comparison of apples and oranges.
-
Anthropic Extracts Interpretable Features from Transformer Layer
By
–
Anthropic’s new research extracts a large number of interpretable features from a one-layer transformer. What does that mean?
The neural networks in large language models show superposition. That means each neuron in the network represents more than one unique feature. -
Google Q&A on Neural Implicit Representations at ICCV2023
By
–
Visit the #ICCV2023 Google booth today at 3:30 PM for a Q&A session on the limitations of employing neural implicit representations for real world applications. Don’t miss the opportunity to learn about Google research contributions toward this goal. pic.twitter.com/a7rY5QjtYN
— Google AI (@GoogleAI) 6 octobre 2023Visit the #ICCV2023 Google booth today at 3:30 PM for a Q&A session on the limitations of employing neural implicit representations for real world applications. Don’t miss the opportunity to learn about Google research contributions toward this goal.
-
Google Researcher Presents Prompt-Tuning Method for Inverse Imaging
By
–
Today at 12:30 PM, visit the #ICCV2023 Google booth to listen to Google Student Researcher, Hyungjin Chung, discuss a new prompt-tuning method that solves inverse imaging problems by using text-to-image latent diffusion models as general priors.
-
PhD Research Excellence Over Social Media Recognition
By
–
Let it fuel your fire! They are probably getting requests for pictures all day long (at ICCV?) Better to do amazing research as a PhD candidate and stand out through a great presentation.
-
CS Undergraduates Transitioning to Computational Neuroscience Research
By
–
Speaking to more and more CS undergraduates who are disillusioned by the current state of AI. On the bright side, I met one that wants to apply to computational neuroscience grad programs. What background do CS majors need to help out in that field?