Chip Placement with Deep Reinforcement Learning (Paper Explained)
The AI Singularity is here! Computers designing new computers! It takes human experts multiple weeks to design new computer chips. What looks like a large game of Tetris is actually a very complex optimization problem. This paper uses Deep Reinforcement Learning to solve this optimization both faster and better than humans.
https://arxiv.org/abs/2004.10746
Abstract:
In this work, we present a learning-based approach to chip placement, one of the most complex and time-consuming stages of the chip design process. Unlike prior methods, our approach has the ability to learn from past experience and improve over time. In particular, as we train over a greater number of chip blocks, our method becomes better at rapidly generating optimized placements for previously unseen chip blocks. To achieve these results, we pose placement as a Reinforcement Learning (RL) problem and train an agent to place the nodes of a chip netlist onto a chip canvas. To enable our RL policy to generalize to unseen blocks, we ground representation learning in the supervised task of predicting placement quality. By designing a neural architecture that can accurately predict reward across a wide variety of netlists and their placements, we are able to generate rich feature embeddings of the input netlists. We then use this architecture as the encoder of our policy and value networks to enable transfer learning. Our objective is to minimize PPA (power, performance, and area), and we show that, in under 6 hours, our method can generate placements that are superhuman or comparable on modern accelerator netlists, whereas existing baselines require human experts in the loop and take several weeks.
Authors: Azalia Mirhoseini, Anna Goldie, Mustafa Yazgan, Joe Jiang, Ebrahim Songhori, Shen Wang, Young-Joon Lee, Eric Johnson, Omkar Pathak, Sungmin Bae, Azade Nazi, Jiwoo Pak, Andy Tong, Kavya Srinivasa, William Hang, Emre Tuncer, Anand Babu, Quoc V. Le, James Laudon, Richard Ho, Roger Carpenter, Jeff Dean
Links:
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.minds.com/ykilcher
Видео Chip Placement with Deep Reinforcement Learning (Paper Explained) канала Yannic Kilcher
https://arxiv.org/abs/2004.10746
Abstract:
In this work, we present a learning-based approach to chip placement, one of the most complex and time-consuming stages of the chip design process. Unlike prior methods, our approach has the ability to learn from past experience and improve over time. In particular, as we train over a greater number of chip blocks, our method becomes better at rapidly generating optimized placements for previously unseen chip blocks. To achieve these results, we pose placement as a Reinforcement Learning (RL) problem and train an agent to place the nodes of a chip netlist onto a chip canvas. To enable our RL policy to generalize to unseen blocks, we ground representation learning in the supervised task of predicting placement quality. By designing a neural architecture that can accurately predict reward across a wide variety of netlists and their placements, we are able to generate rich feature embeddings of the input netlists. We then use this architecture as the encoder of our policy and value networks to enable transfer learning. Our objective is to minimize PPA (power, performance, and area), and we show that, in under 6 hours, our method can generate placements that are superhuman or comparable on modern accelerator netlists, whereas existing baselines require human experts in the loop and take several weeks.
Authors: Azalia Mirhoseini, Anna Goldie, Mustafa Yazgan, Joe Jiang, Ebrahim Songhori, Shen Wang, Young-Joon Lee, Eric Johnson, Omkar Pathak, Sungmin Bae, Azade Nazi, Jiwoo Pak, Andy Tong, Kavya Srinivasa, William Hang, Emre Tuncer, Anand Babu, Quoc V. Le, James Laudon, Richard Ho, Roger Carpenter, Jeff Dean
Links:
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.minds.com/ykilcher
Видео Chip Placement with Deep Reinforcement Learning (Paper Explained) канала Yannic Kilcher
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Reinforcement Learning with Augmented Data (Paper Explained)Decision Transformer: Reinforcement Learning via Sequence Modeling (Research Paper Explained)Keynote: ML for Systems and Chip Design - Azalia Mirhoseini, Senior Research Scientist, Google BrainMIT 6.S094: Introduction to Deep Learning and Self-Driving CarsBYOL: Bootstrap Your Own Latent: A New Approach to Self-Supervised Learning (Paper Explained)I talk to the new Facebook Blender ChatbotGradient Surgery for Multi-Task LearningTraining AI Without Writing A Reward Function, with Reward ModellingHow This Guy Uses A.I. to Create Art | Obsessed | WIREDGoogle's Deep Mind Explained! - Self Learning A.I.The AI Economist: Improving Equality and Productivity with AI-Driven Tax Policies (Paper Explained)Deep Reinforcement Learning-Based Dynamic Scheduling in Smart ManufacturingMIT 6.S091: Introduction to Deep Reinforcement Learning (Deep RL)Fast reinforcement learning with generalized policy updates (Paper Explained)MLP-Mixer: An all-MLP Architecture for Vision (Machine Learning Research Paper Explained)FNet: Mixing Tokens with Fourier Transforms (Machine Learning Research Paper Explained)AI Plays Hide and Seek...Surprises Creators (Bonus Scenes)FixMatch: Simplifying Semi-Supervised Learning with Consistency and ConfidenceThe incredible inventions of intuitive AI | Maurice Conti