VLSI Mask Optimization: From Shallow To Deep Learning Haoyu Yang 1 , Wei Zhong 2 , Yuzhe Ma 1 , Hao Geng 1 , Ran Chen 1 , Wanli Chen 1 , Bei Yu 1 1 The Chinese University of Hong Kong 2 Dalian University of Technology 1 / 22
<latexit sha1_base64="1XOQUpYJ4Navgj2rMp+JfrNOobQ=">AB7nicbVA9TwJBEJ3DL8Qv1NJmI5hYkTsaLElsLDERMIEL2VsW2LC7d9mdMyEXfoSNhcbY+nvs/DcucIWCL5nk5b2ZzMyLEiks+v63V9ja3tndK+6XDg6Pjk/Kp2cdG6eG8TaLZWweI2q5FJq3UaDkj4nhVEWSd6Pp7cLvPnFjRawfcJbwUNGxFiPBKDqpW+2rlKjqoFzxa/4SZJMEOalAjtag/NUfxixVXCOT1Npe4CcYZtSgYJLPS/3U8oSyKR3znqOaKm7DbHnunFw5ZUhGsXGlkSzV3xMZVdbOVOQ6FcWJXfcW4n9eL8XRTZgJnaTINVstGqWSYEwWv5OhMJyhnDlCmRHuVsIm1FCGLqGSCyFYf3mTdOq1wK8F9/VKs5HUYQLuIRrCKABTbiDFrSBwRSe4RXevMR78d69j1VrwctnzuEPvM8fNUyOxg=</latexit> <latexit sha1_base64="1XOQUpYJ4Navgj2rMp+JfrNOobQ=">AB7nicbVA9TwJBEJ3DL8Qv1NJmI5hYkTsaLElsLDERMIEL2VsW2LC7d9mdMyEXfoSNhcbY+nvs/DcucIWCL5nk5b2ZzMyLEiks+v63V9ja3tndK+6XDg6Pjk/Kp2cdG6eG8TaLZWweI2q5FJq3UaDkj4nhVEWSd6Pp7cLvPnFjRawfcJbwUNGxFiPBKDqpW+2rlKjqoFzxa/4SZJMEOalAjtag/NUfxixVXCOT1Npe4CcYZtSgYJLPS/3U8oSyKR3znqOaKm7DbHnunFw5ZUhGsXGlkSzV3xMZVdbOVOQ6FcWJXfcW4n9eL8XRTZgJnaTINVstGqWSYEwWv5OhMJyhnDlCmRHuVsIm1FCGLqGSCyFYf3mTdOq1wK8F9/VKs5HUYQLuIRrCKABTbiDFrSBwRSe4RXevMR78d69j1VrwctnzuEPvM8fNUyOxg=</latexit> <latexit sha1_base64="1XOQUpYJ4Navgj2rMp+JfrNOobQ=">AB7nicbVA9TwJBEJ3DL8Qv1NJmI5hYkTsaLElsLDERMIEL2VsW2LC7d9mdMyEXfoSNhcbY+nvs/DcucIWCL5nk5b2ZzMyLEiks+v63V9ja3tndK+6XDg6Pjk/Kp2cdG6eG8TaLZWweI2q5FJq3UaDkj4nhVEWSd6Pp7cLvPnFjRawfcJbwUNGxFiPBKDqpW+2rlKjqoFzxa/4SZJMEOalAjtag/NUfxixVXCOT1Npe4CcYZtSgYJLPS/3U8oSyKR3znqOaKm7DbHnunFw5ZUhGsXGlkSzV3xMZVdbOVOQ6FcWJXfcW4n9eL8XRTZgJnaTINVstGqWSYEwWv5OhMJyhnDlCmRHuVsIm1FCGLqGSCyFYf3mTdOq1wK8F9/VKs5HUYQLuIRrCKABTbiDFrSBwRSe4RXevMR78d69j1VrwctnzuEPvM8fNUyOxg=</latexit> <latexit sha1_base64="1XOQUpYJ4Navgj2rMp+JfrNOobQ=">AB7nicbVA9TwJBEJ3DL8Qv1NJmI5hYkTsaLElsLDERMIEL2VsW2LC7d9mdMyEXfoSNhcbY+nvs/DcucIWCL5nk5b2ZzMyLEiks+v63V9ja3tndK+6XDg6Pjk/Kp2cdG6eG8TaLZWweI2q5FJq3UaDkj4nhVEWSd6Pp7cLvPnFjRawfcJbwUNGxFiPBKDqpW+2rlKjqoFzxa/4SZJMEOalAjtag/NUfxixVXCOT1Npe4CcYZtSgYJLPS/3U8oSyKR3znqOaKm7DbHnunFw5ZUhGsXGlkSzV3xMZVdbOVOQ6FcWJXfcW4n9eL8XRTZgJnaTINVstGqWSYEwWv5OhMJyhnDlCmRHuVsIm1FCGLqGSCyFYf3mTdOq1wK8F9/VKs5HUYQLuIRrCKABTbiDFrSBwRSe4RXevMR78d69j1VrwctnzuEPvM8fNUyOxg=</latexit> Moore’s Law to Extreme Scaling Moore’s Law Process Technology ( µm <latexit sha1_base64="1XOQUpYJ4Navgj2rMp+JfrNOobQ=">AB7nicbVA9TwJBEJ3DL8Qv1NJmI5hYkTsaLElsLDERMIEL2VsW2LC7d9mdMyEXfoSNhcbY+nvs/DcucIWCL5nk5b2ZzMyLEiks+v63V9ja3tndK+6XDg6Pjk/Kp2cdG6eG8TaLZWweI2q5FJq3UaDkj4nhVEWSd6Pp7cLvPnFjRawfcJbwUNGxFiPBKDqpW+2rlKjqoFzxa/4SZJMEOalAjtag/NUfxixVXCOT1Npe4CcYZtSgYJLPS/3U8oSyKR3znqOaKm7DbHnunFw5ZUhGsXGlkSzV3xMZVdbOVOQ6FcWJXfcW4n9eL8XRTZgJnaTINVstGqWSYEwWv5OhMJyhnDlCmRHuVsIm1FCGLqGSCyFYf3mTdOq1wK8F9/VKs5HUYQLuIRrCKABTbiDFrSBwRSe4RXevMR78d69j1VrwctnzuEPvM8fNUyOxg=</latexit> <latexit sha1_base64="1XOQUpYJ4Navgj2rMp+JfrNOobQ=">AB7nicbVA9TwJBEJ3DL8Qv1NJmI5hYkTsaLElsLDERMIEL2VsW2LC7d9mdMyEXfoSNhcbY+nvs/DcucIWCL5nk5b2ZzMyLEiks+v63V9ja3tndK+6XDg6Pjk/Kp2cdG6eG8TaLZWweI2q5FJq3UaDkj4nhVEWSd6Pp7cLvPnFjRawfcJbwUNGxFiPBKDqpW+2rlKjqoFzxa/4SZJMEOalAjtag/NUfxixVXCOT1Npe4CcYZtSgYJLPS/3U8oSyKR3znqOaKm7DbHnunFw5ZUhGsXGlkSzV3xMZVdbOVOQ6FcWJXfcW4n9eL8XRTZgJnaTINVstGqWSYEwWv5OhMJyhnDlCmRHuVsIm1FCGLqGSCyFYf3mTdOq1wK8F9/VKs5HUYQLuIRrCKABTbiDFrSBwRSe4RXevMR78d69j1VrwctnzuEPvM8fNUyOxg=</latexit> <latexit sha1_base64="1XOQUpYJ4Navgj2rMp+JfrNOobQ=">AB7nicbVA9TwJBEJ3DL8Qv1NJmI5hYkTsaLElsLDERMIEL2VsW2LC7d9mdMyEXfoSNhcbY+nvs/DcucIWCL5nk5b2ZzMyLEiks+v63V9ja3tndK+6XDg6Pjk/Kp2cdG6eG8TaLZWweI2q5FJq3UaDkj4nhVEWSd6Pp7cLvPnFjRawfcJbwUNGxFiPBKDqpW+2rlKjqoFzxa/4SZJMEOalAjtag/NUfxixVXCOT1Npe4CcYZtSgYJLPS/3U8oSyKR3znqOaKm7DbHnunFw5ZUhGsXGlkSzV3xMZVdbOVOQ6FcWJXfcW4n9eL8XRTZgJnaTINVstGqWSYEwWv5OhMJyhnDlCmRHuVsIm1FCGLqGSCyFYf3mTdOq1wK8F9/VKs5HUYQLuIRrCKABTbiDFrSBwRSe4RXevMR78d69j1VrwctnzuEPvM8fNUyOxg=</latexit> <latexit sha1_base64="1XOQUpYJ4Navgj2rMp+JfrNOobQ=">AB7nicbVA9TwJBEJ3DL8Qv1NJmI5hYkTsaLElsLDERMIEL2VsW2LC7d9mdMyEXfoSNhcbY+nvs/DcucIWCL5nk5b2ZzMyLEiks+v63V9ja3tndK+6XDg6Pjk/Kp2cdG6eG8TaLZWweI2q5FJq3UaDkj4nhVEWSd6Pp7cLvPnFjRawfcJbwUNGxFiPBKDqpW+2rlKjqoFzxa/4SZJMEOalAjtag/NUfxixVXCOT1Npe4CcYZtSgYJLPS/3U8oSyKR3znqOaKm7DbHnunFw5ZUhGsXGlkSzV3xMZVdbOVOQ6FcWJXfcW4n9eL8XRTZgJnaTINVstGqWSYEwWv5OhMJyhnDlCmRHuVsIm1FCGLqGSCyFYf3mTdOq1wK8F9/VKs5HUYQLuIRrCKABTbiDFrSBwRSe4RXevMR78d69j1VrwctnzuEPvM8fNUyOxg=</latexit> µm ) 10 1 0.1 0.01 A12 10,000,000,000 A7 Core i7 1,000,000,000 Core 2 Duo A10 A11 Intel Microprocessors Number of Transistors per Integrated Circuit Pentium 4 Apple Microprocessors 100,000,000 Pentium II 10,000,000 486 Pentium 1,000,000 386 Invention of the 286 100,000 Transistor 8086 Doubles every 2.1 yrs 4004 10,000 1,000 100 10 1 1940 1950 1960 1970 1980 1990 2000 2010 2020 Year 2 / 22
Challenge 1: Failure (Hotspot) Detection Hotspot on Wafer Pre-OPC Layout Post-OPC Mask Ra#o%of%lithography%simula#on%#me% Required(computa/onal( /me(reduc/on! � (normalized%by%40nm%node)% ◮ RET: OPC, SRAF, MPL ◮ Still hotspot: low fidelity patterns ◮ Simulations: extremely CPU intensive Technology%node � 3 / 22
Challenge 2: Optical Proximity Correction (OPC) Wafer Mask Design target without OPC with OPC 4 / 22
Why Deep Learning? ◮ Feature Crafting v.s. Feature Learning Although prior knowledge is considered during manually feature design, information loss is inevitable. Feature learned from mass dataset is more reliable. ◮ Scalability With shrinking down circuit feature size, mask layout becomes more complicated. Deep learning has the potential to handle ultra-large-scale instances while traditional machine learning may suffer from performance degradation. ◮ Mature Libraries 5 / 22
Outline Hotspot Detection via Machine Learning OPC via Machine Learning Heterogeneous OPC 6 / 22
Outline Hotspot Detection via Machine Learning OPC via Machine Learning Heterogeneous OPC 7 / 22
Hotspot Detection Hierarchy Sampling Increasing Hotspot Detection verification accuracy Lithography Simulation (Relative) CPU runtime at each level ◮ Sampling (DRC Checking): scan and rule check each region ◮ Hotspot Detection : verify the sampled regions and report potential hotspots ◮ Lithography Simulation : final verification on the reported hotspots 7 / 22
Early Study of DNN-based Hotspot Detector ∗ ◮ Total 21 layers with 13 convolution layers and 5 pooling layers. ◮ A ReLU is applied after each convolution layer. C1 C2-1C2-2 C2-3 P1 P2 C3-1 C3-2 C3-3 P3 C4-1 C4-2 C4-3 P4 C5-1 C5-2 C5-3 P5 … Hotspot … 16x16x32 Non-Hotspot 32x32x32 32x32x32 64x64x32 128x128x16 64x64x16 128x128x8 256x256x8 256x256x4 512x512x4 512 2048 ∗ Haoyu Yang, Luyang Luo, et al. (2017). “Imbalance aware lithography hotspot detection: a deep learning approach”. In: JM3 16.3, p. 033504. 8 / 22
What Does Deep Learning Learn? Origin Pool1 Pool2 Pool3 Pool4 Pool5 9 / 22
The Biased Learning Algorithm [DAC’17] † Shift-Boundary Bias Training Set Update ε MGD: y h =[0,1] end-to-end 4 , 000 y n =[1- ε , ε ] training False Alarm No 3 , 000 Stop Criteria Yes Trained Model 2 , 000 80 82 84 86 88 90 Accuracy (%) † Haoyu Yang, Jing Su, et al. (2017). “Layout Hotspot Detection with Feature Tensor Generation and Deep Biased Learning”. In: Proc. DAC , 62:1–62:6. 10 / 22
Optimizing AUC [ASPDAC’19] ‡ The AUC objective: � � �� � N + � N − 1 � x + � L Φ ( f ) = j = 1 Φ f − f . x − N + N − i = 1 i j Approximation candidates: PSL Φ PSL ( z ) = ( 1 − z ) 2 PHL Φ PHL ( z ) = max( 1 − z , 0 ) PLL Φ PLL ( z ) = log( 1 + exp( − β z )) � − ( z − γ ) p , if z > γ R Φ R ∗ ( z ) = 0 , otherwise ‡ Wei Ye et al. (2019). “LithoROC: lithography hotspot detection with explicit ROC optimization”. In: Proc. ASPDAC , pp. 292–298. 11 / 22
Conventional Clip based Solution Hotspot … Conventional Hotspot Detector Non- Hotspot Clips Region ◮ A binary classification problem. ◮ Scan over whole region. ◮ Single stage detector. ◮ Scanning is time consuming and single stage is not robust to false alarm. 12 / 22
Region based approach [DAC’19] Hotspot Core Region-based Hotspot Detector Feature Extraction Clip Proposal Network Refinement Region ◮ Learning what and where is hotspot at same time. ◮ Classification Problem -> Classification & Regression Problem. Ran Chen et al. (2019). “Faster Region-based Hotspot Detection”. In: Proc. DAC , 146:1–146:6. 13 / 22
Outline Hotspot Detection via Machine Learning OPC via Machine Learning Heterogeneous OPC 14 / 22
OPC Previous Work Classic OPC Machine Learning OPC ◮ Model/Rule-based OPC [Matsunawa+,JM3’16][Choi+,SPIE’16] [Cobb+,SPIE’02][Kuang+,DATE’15] [Xu+,ISPD’16][Shim+,APCCAS’16] [Awad+,DAC’16][Su+,ICCAD’16] 1. Edge fragmentation; 1. Fragmentation of shape edges; 2. Feature extraction; 2. Move fragments for better printability. 3. Model training. ◮ Inverse Lithography [Pang+,SPIE’05][Gao+,DAC’14] [Poonawala+,TIP’07][Ma+,ICCAD’17] 1. Efficient model that maps mask to aerial image; 2. Continuously update mask through descending the gradient of contour error. 14 / 22
Recommend
More recommend
