filmov
tv
Chip Placement with Deep Reinforcement Learning (Paper Explained)
Показать описание
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.
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:
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:
0:27:27
Chip Placement with Deep Reinforcement Learning (Paper Explained)
0:35:41
Chip Placement with Deep Reinforcement Learning | Paper Explained
0:50:19
CHIPS Alliance - Learning To Play the Game of Macro Placement with Deep Reinforcement Learning
0:28:27
Chip Placement with Deep Reinforcement Learning #2
0:23:16
Chip Placement with Deep Reinforcement Learning #1
0:09:33
Chip floorplanning with deep reinforcement learning
0:34:12
Chip Placement with Deep Reinforcement Learning 2004 10746v1 05
0:31:52
Chip Placement with Deep Reinforcement Learning #4
0:23:10
Chip Placement with Deep Reinforcement Learning #3
0:57:26
Reinforcement Learning for Hardware Design feat. Anna Goldie | Stanford MLSys Seminar Episode 14
0:35:26
M4LA 2021 - Anna Goldie - Model Parallelism and Chip Floorplanning with Deep Reinforcement Learning
0:29:41
Thinking While Moving: Deep Reinforcement Learning with Concurrent Control
0:09:20
AI Learns Parallel Parking - Deep Reinforcement Learning
0:01:28
Neuromorphic Accelerator Test-chip for Online Reinforcement Learning
0:22:15
Introduction to Deep Reinforcement Learning | Deep RL Course
0:05:20
BayLearn 2020: Optimizing Memory Placement using Evolutionary Graph Reinforcement Learning
3:16:35
IEEE Seasonal School: Deep / Reinforcement Learning (2022-11-04)
0:32:01
Offline Deep Reinforcement Learning Algorithms
0:10:38
Deep Reinforcement Learning: Field Development Optimization | Paper Explained
0:22:11
Device Placement Optimisation with Reinforcement Learning
0:00:31
Predicting Mid-Air Interaction Movements and Fatigue Using Deep Reinforcement Learning
0:14:05
Deep Reinforcement Learning at the Edge of Statistical Precipice | Outstanding Paper | NeurIPS'...
0:58:02
Solving Optimization Problems in Systems and Chip Design: Google Brain Research
0:21:35
Device Placement Optimization with Reinforcement Learning
Комментарии