YunpyoAn (@yunpyoan) 's Twitter Profile
YunpyoAn

@yunpyoan

Ph.D. Candidate in Artificial intelligence at UNIST
B.S. at UNIST, major CompSci /
I usually post Korean...

ID: 1266287914524143616

linkhttp://raon1123.github.io calendar_today29-05-2020 08:39:06

12,12K Tweet

576 Takipçi

1,1K Takip Edilen

Ernest Ryu (@ernestryu) 's Twitter Profile Photo

New lecture recordings on RL+LLM! 📺 This spring, I gave a lecture series titled **Reinforcement Learning of Large Language Models**. I have decided to re-record these lectures and share them on YouTube. (1/7)

Sergey Levine (@svlevine) 's Twitter Profile Photo

Action chunking is a great idea in robotics: by getting a model to produce a short sequence of actions, it _just works better_ for some mysterious reason. Now it turns out this can help in RL too, and it's a bit clearer why: action chunks help explore and help with backups. 🧵👇

Action chunking is a great idea in robotics: by getting a model to produce a short sequence of actions, it _just works better_ for some mysterious reason. Now it turns out this can help in RL too, and it's a bit clearer why: action chunks help explore and help with backups. 🧵👇
Paul Zhou (@zhiyuan_zhou_) 's Twitter Profile Photo

Action chunking works really well in imitation learning, and is essential to learning good BC policies in robotics. Can/should we apply the same idea in RL? We find that RL in the action chunk space, when done right (we call it ✨Q-chunking ✨), can be highly efficient🧵👇

Action chunking works really well in imitation learning, and is essential to learning good BC policies in robotics. Can/should we apply the same idea in RL? 

We find that RL in the action chunk space, when done right (we call it ✨Q-chunking ✨), can be highly efficient🧵👇
YunpyoAn (@yunpyoan) 's Twitter Profile Photo

세계유산에서 가장 가까운 과기원 (아마도) "울산과학기술원" == 학교 뒤에 호수 (안에 말고) 거기 근처에 있다고 들음. 다만 빙 둘러 가야함.

YunpyoAn (@yunpyoan) 's Twitter Profile Photo

회사든 학교든 높은 연봉이 이직 사유라면, 동일한 이유로 퇴직할 가능성이 높다고 생각한다. 언젠가는 연봉 상승의 한계를 마주할 것이고, 퇴직하게 되는 것이 자연스러운 흐름이다.

YunpyoAn (@yunpyoan) 's Twitter Profile Photo

흠... 대나무숲이 있는 이유를 알거 같음. 뭔가 진짜 데인 경험을 직간접적으로 알고 있는데, 책 한 권 나올 수 있을거 같음. 요약 악!

상일 (@sioum) 's Twitter Profile Photo

"UNIST는 신임 교원을 위한 초기 정착연구비를 최대 3억원까지 지원하고, 특훈교수로 선정된 교수에게는 정교수 대비 200% 이상의 임금을 지급하는 등 특급 대우를 하고 있다." mk.co.kr/news/society/1…