exploiting the unused part of the brain - caltech group at ... · 11/16/16 sc16, brain inspired...
Post on 09-May-2020
1 Views
Preview:
TRANSCRIPT
Exploiting the UnusedPart of the Brain
Deep Learning and Emerging TechnologyFor High Energy Physics
Jean-Roch Vlimant
A 10 Megapixel Camera
CMS 100 Megapixel Camera
CMS Detector
CMS Readout
Highly heterogeneous system Raw data is 100M channelssampled every 25 ns : 1Pb/s50EB per day in readout and
online processing.
Event Filtering
From Big Data to Smart Datawith ultra fast decision
1000 Gb/s1000
Gb
/s
105 H
z
40 M
Hz
1-3
kHz
L1 HLT
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 7
Why Deep Learning● LHC Data Processing may use deep learning methods in many aspects (attend
other relevant talk at the Caltech booth)● Large volume of data and simulated data to analyze● Several class of LHC Data analysis make use of classifier for signal versus
background discrimination.✔ Use of BDT on high level features✔ Increasing use of MLP-like deep neural net
● Deep learning has delivered super-human performance at certain class of tasks(computer vision, speech recognition, ...)
✔ Use of convolutional neural net, recurrent topologies, long-short-term-memorycells, ...
● Deep learning has advantage at training on “raw” data➢ Several levels of data distillation in LHC data processing
● Neural net computation is highly parallelizable
➔ Going beyond fully connected networks with advanced deep neural net topologies➢ Multi-classification of LHC events from particle-level information➢ Charged particle tracking with recurrent and convolutional topologies➢ Particle identification and energy regression in highly granular future
calorimeter➢ ...
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 8
Advanced Machine Learningand Deep Learning
(my selection)
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 9
Machine Learning in a Nutshell
Balazs Kegl, CERN 2014
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 10
Scene Labeling
Karpathy, Fei-Fei, CVPR 2015
● Create a description of images
➢ Generate a decay process description from collisionrepresentation, with application to triggers
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 11
Scenery Interpretation
Farabet et al. ICML 2012, PAMI 2013
● Group and classify what each pixel belongs to● Real-time video processing with deep learning
➢ Multiple applications to pileup mitigation, objectidentification, tracking. All from “raw data”
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 12
Attention Learning
● Identify people from faces with multiple attentionfilters
➢ Object identification, noise subtraction, ...
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 13
Text Processing
● Question and Answer machine, languagetranslation, semantic arithmetic, ...
➢ Can the raw data of detector be interpreted astexts and translated into physics descriptions ?
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 14
Embedded Symmetries
T.S. Cohen, M. Welling ICML2016
p4m group
p4 group
● Introduction of convolutionallayers was a ground-breaking advancement
● Research on embedding morefundamental symmetries intoneural nets
● Symmetries operate on thedata or internalrepresentation of data
● Next is to implementsymmetries of physics to buildphysics-specific NN
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 15
Toolkit and Services
● Lots of libraries out there, several key components in each majorlanguages. Lots of big-data analytics services offered
● Common theme of going for spark-hdfs support➢ Question of having in-house software or embracing external
libraries is very much alive
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 16
Application to Intensity andEnergy Frontiers
(a selected few)
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 17
NOVA Event Classification
Slides on Paolo
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 18
Particle Jet Identification
Neural net
Train
W to QCDdiscrimination
W tagger arXiv: 1511.05190, Oliveira, Kagan, Mackey, Nachman, Schwartzman
Top Tagger arXiv: 1501.05968 Almeida, Backovic, Cliche, Lee, Perelstein
W to QCDdiscrimination
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 19
3D Calorimetry Imaging
100GeV Photon 100GeV Pi0
≠LCD Calorimeter configurationhttp://lcd.web.cern.ch5x5 mm Pixel calorimeter28 layer deep for Ecal70 layer deep for Hcal
Photon and pion particle gunClassification, regression andcombined models
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 20
Irregular Geometry Challenge
Hexagonal cells
Projective Geometry
Variable Depth Segmentation
The images we are dealing with arenot as regular as standard images.Need for specific new treatment andmethods to feed neural nets
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 21
Collision Event Classification
● Full event classification using reconstructed particle 4-vectors● Recurrent neural nets, Long short term memory cells● Dedicated layer with Lorentz boosting● Step toward event classification with lower level data : low
level feature as opposed to analysis level variables
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 22
Ordering ChallengeText have naturalorder. RNN/LSTMcan correlate theinformation tointernalrepresentation
There isunderlying order incollision events.Smeared throughtiming resolution.No natural order in observable
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 23
Charged Particle Tracking
● Perfect example of pattern recognition● Data sparsity is not common in image processing● Several angles to tackle the problem. Deep Kalman filter,
RNN to learn dynamics, sparse image processing, ...● Kaggle challenge in preparation
https://indico.hep.caltech.edu/indico/event/102
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 24
Challenge of Tracking
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 25
HEP Trk.Xhttps://heptrkx.github.io/
● LBL, FNAL, Caltech consortium sponsored by the DOE● Preparation of simulated data
➢ Accurate, fast and light● Explore new approaches to charged particle tracking using
advanced pattern recognition techniques➢ Recurrent neural nets in learning track kinematics➢ Convolutional neural nets for pattern recognition➢ Hough-like transformation from hit position space to track
parameter space➢ Advanced Kalman filter parallelization➢ ... exploration has started
● Tracking competition (a la kaggle) in preparation
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 26
Other Applications
● Outliers selection● Anomaly detection● Data quality automation● Detector control● Experiment control● Data popularity prediction● Computing grid control● Denoising with auto-encoder● Fast simulation ● ...
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 27
Accelerating and EmergingTechnologies
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 28
Supermicro Server
Caltech Servers● 2 compute nodes :
Intel® Xeon® CPU E5-2697 v3 @ 2.60GHz processors pernode (28 cores per CPU) with 8 NVIDIA® Pascal® GTX 1080
● Theoretical Peak Performance :80 Tflops
➔ Theano, Tensorflow, Keras➔ MPI training➔ Spearmint hyper-optimization
Many thanks to our partners Nvidia and Supermicro
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 29
ALCF
Cooley● 126 compute nodes :
Two 2.4 GHz Intel®Haswell® E5-2620 v3processors per node (6 coresper CPU, 12 cores total) andNVIDIA® Tesla® K80
● Theoretical Peak Performance :293 Tflops
➔ Development Project with 8kcore hours
➔ Theano, Tensorflow, Keras➔ MPI ready➔ Spearmint experimental
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 30
CSCS
Piz Daint● 5272 compute nodes :
Intel® Xeon® E5-2670 @2.60GHz (8 cores, 16 virtualcores with hyperthreadingenabled, 32GB RAM) andNVIDIA® Tesla® K20X
● Theoretical Peak Performance :7.787 Pfops
● Scratch capacity : 2.7 PB
➔ Development Project with 36kcore hours
➔ Theano, Tensorflow, Keras➔ MPI ready➔ Spark experimental stage
OLCF
Titan● 18688 compute nodes :
2.2GHz AMD® Opteron®6274 processors per node (16cores per CPU) and NVIDIA®Tesla® K20X
● Theoretical Peak Performance :20 Pflops
➔ Allocation in preparation
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 32
Distributed Learning
● Deep learning with elastic averaging SGDhttps://arxiv.org/abs/1412.6651
● Revisiting Distributed Synchronous SGDhttps://arxiv.org/abs/1604.00981
● Implementation with Spark and MPI for the Kerasframework https://keras.io/
➔ https://github.com/JoeriHermans/dist-keras ➔ https://github.com/duanders/mpi_learn
Titan X
GTX 1080
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 33
Performance
● Ran on the supermicro 8GPU server here at SC16● Normalized to 2 GPU performance point
➔ Please add a factor 2x
Linear speedup withnumber of workers
Saturation in co-scheduling GPU (2 workers on the same GPU)
7x max speedup
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 34
Performance
● Ran on ALCF Cooley 126 GPU cluster● Normalized to 2 GPU performance point
➔ Please add a factor 2x
14x max speedup
Non-linear speedup withnumber of GPU
Linear speedup withnumber of GPU
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 35
Applicability
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 36
Training vs Inference
● GPUs are the workhorse forparallel computing
● Enable training large models, withlarge dataset
● Deep learning facility clusters
● Emergence of smaller GPU● Not dedicated to training● Strike the balance between
Tflops/$ for inference● Deployment on the grid
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 37
Neuromorphic Hardware
http://www.nature.com/articles/srep14730
● Implementing plasticity in hardware ● Process signal from detector and adapt tocategories of pattern (unsupervised)
● Post-classified from data analysis or rate throttling➢ NCCR consortium assembling to develop thistechnology further, with our use case in mind
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 38
Cognitive Computing
● Spiking neural net as processing units : ➔Cognitive Computing Processing Unit : CCPU
● Adopt a new programming scheme, translateexisting software
● See Rebecca Carney's talk for more details
11/16/16 SC16, Brain Inspired Technologies, J.-R. Vlimant 39
Summary
Impressive achievement and promise ofmodern machine learning and deep learning
From realistic to speculative applicability tofield of High Energy and Frontier Physics
Emerging tools and technology to embrace
Thanks to our sponsors
top related