- A Deep Reinforcement learning approach to chip floorplanning automatically generates chip floorplans in under six hours that are superior to those produced by humans in all key metrics, including power consumption, performance and chip area. These layouts were used in Google's latest AI accelerator. https://www.nature.com/articles/s41586-021-03544-w 3 comments technology
- Scientists at Google described an AI-based approach to chip design that could learn from past experience and improve over time, becoming better at generating architectures for unseen components. It completed designs in under six hours on average, faster than the weeks it takes human experts. https://www.nature.com/articles/s41586-021-03544-w 3 comments science
Linking pages
- How AlphaChip transformed computer chip design - Google DeepMind https://deepmind.google/discover/blog/how-alphachip-transformed-computer-chip-design/ 212 comments
- AI is Ushering In a New Scientific Revolution https://thegradient.pub/ai-scientific-revolution/ 143 comments
- Alan’s conservative countdown to AGI – Dr Alan D. Thompson – Life Architect https://lifearchitect.ai/agi/ 73 comments
- Tension Inside Google Over a Fired AI Researcher’s Conduct | WIRED https://www.wired.com/story/google-brain-ai-researcher-fired-tension/ 23 comments
- Need to Fit Billions of Transistors on a Chip? Let AI Do It | WIRED https://www.wired.com/story/fit-billions-transistors-chip-let-ai-do/ 16 comments
- Google is using AI to design its next generation of AI chips more quickly than humans can - The Verge https://www.theverge.com/2021/6/10/22527476/google-machine-learning-chip-design-tpu-floorplanning 15 comments
- Recent Advances in Deep Learning for Routing Problems | Chaitanya K. Joshi https://www.chaitjo.com/post/deep-learning-for-routing-problems/ 3 comments
- Google used reinforcement learning to design next-gen AI accelerator chips | VentureBeat https://venturebeat.com/2021/06/09/google-used-reinforcement-learning-to-design-next-gen-ai-accelerator-chips/ 2 comments
- Bursting the Bubble: A Case Against Reinforcement Learning for Macro Placement | by MLWhistleDown | Medium https://medium.com/@mlgit_70611/bursting-the-bubble-a-case-against-reinforcement-learning-for-macro-placement-1c0d014c4143 1 comment
- Nature flags doubts over Google AI study, pulls commentary – Retraction Watch https://retractionwatch.com/2023/09/26/nature-flags-doubts-over-google-ai-study-pulls-commentary/ 1 comment
- GitHub - google-research/circuit_training https://github.com/google-research/circuit_training 0 comments
- Gradient Update #3: New in Reinforcement Learning - Chip Design and Transformers https://thegradientpub.substack.com/p/update-3-new-in-reinforcement-learning 0 comments
- Spotting the Next Big Thing in Semiconductors https://semiliterate.substack.com/p/spotting-the-next-big-thing-in-semiconductors 0 comments
- Now that computers connect us all, for better and worse, what’s next? https://www.sciencenews.org/article/computer-science-history-ethics-future-robots-ai 0 comments
- Google faces internal battle over research on AI to speed chip design | Reuters https://www.reuters.com/technology/google-faces-internal-battle-over-research-ai-speed-chip-design-2022-05-03/ 0 comments
- Google's Controversial AI Chip Paper Under Scrutiny Again https://www.hpcwire.com/2023/10/03/googles-controversial-ai-chip-paper-under-scrutiny-again/ 0 comments
Related searches:
Search whole site: site:nature.com
Search title: A graph placement methodology for fast chip design | Nature
See how to search.