#JEPA
World Model 삼국지: Fei-Fei Li, LeCun, DeepMind가 만드는 세 가지 미래
Fei-Fei Li, Yann LeCun, DeepMind가 World Model이라는 같은 용어로 만드는 전혀 다른 세 가지 AI 미래. 3D 자산, 예측 엔진, 시뮬레이터의 차이를 명확히 설명합니다.
LLM-JEPA: Phương pháp đào tạo LLM mới, áp dụng kiến trúc JEPA từ thị giác máy tính, đã ra đời! Nó giúp các Mô hình Ngôn ngữ Lớn vượt trội hơn đáng kể so với đào tạo tiêu chuẩn và chống overfitting hiệu quả, được kiểm chứng trên Llama3, Gemma2... Tuy nhiên, cần lưu ý về 2 siêu tham số bổ sung và chi phí tính toán tăng gấp đôi.
#LLM #JEPA #AI #DeepLearning #MachineLearning #MôHìnhNgônNgữLớn #TríTuệNhânTạo
https://www.reddit.com/r/LocalLLaMA/comments/1o4av71/llmjepa_large_language_models_meet_join
Self-supervised learning, JEPA, world models, and the future of AI [video]
https://www.youtube.com/watch?v=yUmDRxV0krg
#HackerNews #SelfSupervisedLearning #JEPA #WorldModels #FutureOfAI #AIResearch
@yann-lecun.bsky.social, Chief AI Scientist at #Meta, shared that Joint Embedding Predictive Architecture (JEPA) model is the next step toward his vision for advanced machine intelligence (AMI) - possibly in the next 5 years! #MITGenAISymposium #MGAIC #JEPA
#ITByte: The #JEPA (Joint-Embedding Predictive Architecture) AI model is a significant development in self-supervised learning, spearheaded by Meta AI and heavily influenced by Yann LeCun, Meta's Chief AI Scientist.
https://knowledgezone.co.in/posts/Joint-Embedding-Predictive-Architecture-68597054a53194f0fea8f83d
Self-Supervised Learning from Images with JEPA
https://arxiv.org/abs/2301.08243
#HackerNews #SelfSupervisedLearning #JEPA #ImageProcessing #MachineLearning #AIResearch #ComputerVision
"(...) Meta confirms open source language model Llama 3 coming next month
https://kinews24.de/meta-bestaetigt-open-source-sprachmodell-llama-3/
#ArtificialIntelligence #AI #Llama3 #OpenSource #Meta #future #JEPA (...)"
Meta bestätigt Open-Source-Sprachmodell Llama 3 kommt im nächsten Monat
#KuenstlicheIntelligenz #artificialintelligence #KI #AI #Llama3 #OpenSource #Meta #Zukunft #JEPA
https://kinews24.de/meta-bestaetigt-open-source-sprachmodell-llama-3/
〝One significant shortcoming of generative methods, however, is that the model tries to fill-in every bit of missing information, even though the world is inherently unpredictable. As a result, generative methods may be prone to mistakes a person would never make because they focus too much on irrelevant details instead of capturing high-level predictable concepts.〞
#AI #JEPA #Meta
https://ai.meta.com/blog/yann-lecun-ai-model-i-jepa/
#Meta présente #JEPA une #IA #AI qui s'approche de l'intelligence humaine
https://www.01net.com/actualites/mieux-chatgpt-meta-presente-ia-approche-intelligence-humaine.html
Dr Yann LeCun setting the stage for Joint Embedding Predictive Architecture in AI at the Institute for Experiential AI at Northeastern University
-
#Boston #AI #technology #math #abstraction #NortheasternUniversity #YannLeCun #JEPA #machineLearning