-
Notifications
You must be signed in to change notification settings - Fork 40
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
[20220612] Weekly AI ArXiv 만담 - 54회차 #54
Comments
Arxiv (audio and speech processing)
|
Multi-Game Decision Transformers
|
AI's New Frontier 영국 Economist지에서 인공지능과 Foundational Model에 대한 특집을 이번주에 냈습니다. 기술전문 신문이 아님에도 불구하고 최근에 나온 GATO 모델 등에 대해서 매우 자세하고 논의하고 일반인의 입장에서도 Foundation model 및 자연어 모델, DALL-E 등 초거대 모델과 이로 인해 발생하는 이슈 등에 대해 매우 잘 정리했기 때문에 비전문가에게 분야 및 이슈를 소개하는데 매우 좋은 기사라고 생각됩니다. 또한, 저명한 인지과학자이신 Douglas Hofstadter의 GPT3와의 adversarial한 대화를 통해 아직까지도 인공지능이 겉으로는 세계에 대해 이해를 하고 있는 것으로 보일 수 있지만 진정한 World Model을 가지지 못하는 것을 보이는데 최근 이슈가 된 AI 모델과 의식의 관계에 대해 이미 많이 고민을 해본 사람에게도 새로운 각도를 제시해주어 도움이 될 수 있다고 생각합니다. 참조를 위해 이번 주 기사 링크를 공유해드립니다. |
An on-chip photonic deep neural network for image classification Quantum advantage in learning from experiments 딥러닝에 양자 컴퓨터를 사용하는 논문이 각각 Nature와 Science에 accept되었습니다. Beyond the Imitation Game: Collaborative benchmark for measuring and extrapolating the capabilities of language models |
IBM Global AI Adoption Index 2022https://www.ibm.com/watson/resources/ai-adoption
"산업계 AI 활용률 2.5%뿐… 옥석 가려 R&D 집중해야" [AI World 2022]]https://www.fnnews.com/news/202206121752200319
Many researchers were not compliant with their published data sharing statement: mixed-methods studyhttps://linkinghub.elsevier.com/retrieve/pii/S089543562200141X
|
Elucidating the Design Space of Diffusion-Based Generative Models | arxiv
정말 케라스 스타일 그대로 많은 실험과 뛰어난 직감을 가지고 모델을 재구성한 듯 합니다.
|
News
ArXiv
Revisiting the "Video" in Video-Language Understanding
On Data Scaling in Masked Image Modeling
흥미있는 연구
The text was updated successfully, but these errors were encountered: