Pinned
I have so much fun writing this position with some of the most amaaazing people in robotics!
Have a look at it here: paper.motoniq.ai
#AI #MachineLearning #Robotics
Haitham Bou Ammar
4,525 posts
Haitham Bou Ammar
@hbouammar
RL team leader @Noah'a Ark UK H. Assistant Professor @UCL | Ex-@Princeton, Upenn (thou/thine). Co-Founder. Come Collab 🤙🏻🤙🏻
- Bye bye black boxes ? Have you seen this stuff ??? Mind blowing 🎉🎉🎉 #ai #MachineLearning #datafestafrica2022 #DataScience
- Love this work by Calvin Lou. Neat to the point and just a lovely intro to diffusions while being rigorous!! #AI #MachineLearning @GoogleAI @DeepMind @GoogleBrain @GoogleResearch 🎉🎉🤩🤩🤩🤩🤩🤩🤩 Now, we can understand all the mathsss 🤩🤩
- With this hype around LLMs, I am seeing tons of well respected professors publishing half (if not quarter) - baked work just to keep up with the competition! I will advice them as they did advise me once during my PhD. This has to stop, focus on quality and not quantity… I feel
- This is the first paper I read that makes a bit of sense for reasoning. A must-read frmpv. #AI
- I really like those types of books. Have a read through this if you are new to the topic: ics.uci.edu/~dechter/cours… #AI #MachineLearning
- This is deffo one of the best papers I have read in a while! @UniofOxford @yaringal #AI #MachineLearning
- This is a very very very very very cool paper ! Must read! @DeepMind @GoogleAI @mcgillu #ai #MachineLearning #DataScience
- Bias-variance trade-off is one of the most important aspects of #MachineLearning. Don't forget your #algorithmic implementation should strike a balance. #ArtificialIntelligence #BigData #Python #learning #teaching #TensorFlow #PyTorch #DataScience #informationtechnology #math
- 🫠 Yo, after 1 year of really hard work, we have made a new cheap robot - 14K USD bimanual robot. It could do VR-teleoperation, imitation learning, connect with DeepSeek and VLMs to embodied planning tasks. The cool part is we try to make life easier for non-experts by creating
00:00 - So yeah AGI from @OpenAI seems to be Q-learning related. If you want to know what it is here's the Q-learning update rule with a snarky comment :P There are other resources, read about it from @huggingface that inspired me to make this pic: huggingface.co/blog/deep-rl-q…
- I have been reading this paper from @GoogleDeepMind. It is amazing! I like their Monte-Carlo-like collection of data to train reward processes better. I also love the idea of using a binary search to cut the state of the tree (so to say)—it all goes back to the open
