LHC beam mode classification

Slides:



Advertisements
Similar presentations
Artificial Neural Networks (1)
Advertisements

DS Quench TEST 2  MOTIVATION and METHOD: 1. Achieve 500kW on beam 1 – TCP7 collimators.(so far 500kW with beam 2 and 235kW over 1s with beam 1 were reached.
Deep Learning Neural Network with Memory (1)
LHC Collimation Working Group – 20 February 2012 Collimator Setup Software in 2012 G. Valentino R. W. Assmann, S. Redaelli and N. Sammut.
Improving Collimator Setup Efficiency LHC Beam Operation Committee, G. Valentino, R.W. Assmann, R. Bruce, F. Burkart, M. Cauchi, D. Deboy, S.
GPGPU Performance and Power Estimation Using Machine Learning Gene Wu – UT Austin Joseph Greathouse – AMD Research Alexander Lyashevsky – AMD Research.
Predicting the dropouts rate of online course using LSTM method
Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting 卷积LSTM网络:利用机器学习预测短期降雨 施行健 香港科技大学 VALSE 2016/03/23.
Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation EMNLP’14 paper by Kyunghyun Cho, et al.
Neural networks and support vector machines
Welcome deep loria !.
Deep Learning RUSSIR 2017 – Day 3
Unsupervised Learning of Video Representations using LSTMs
RNNs: An example applied to the prediction task
CS 388: Natural Language Processing: LSTM Recurrent Neural Networks
CS 4501: Introduction to Computer Vision Computer Vision + Natural Language Connelly Barnes Some slides from Fei-Fei Li / Andrej Karpathy / Justin Johnson.
Environment Generation with GANs
Speaker Classification through Deep Learning
Computer Science and Engineering, Seoul National University
Convolutional Neural Fabrics by Shreyas Saxena, Jakob Verbeek
Recurrent Neural Networks for Natural Language Processing
Neural Machine Translation by Jointly Learning to Align and Translate
Show and Tell: A Neural Image Caption Generator (CVPR 2015)
A Hierarchical Model of Reviews for Aspect-based Sentiment Analysis
Deep Learning: Model Summary
ICS 491 Big Data Analytics Fall 2017 Deep Learning
Classification with Perceptrons Reading:
Intelligent Information System Lab
Intro to NLP and Deep Learning
Policy Compression for MDPs
Master’s Thesis defense Ming Du Advisor: Dr. Yi Shang
RNNs: Going Beyond the SRN in Language Prediction
A critical review of RNN for sequence learning Zachary C
Grid Long Short-Term Memory
Advanced Artificial Intelligence
Image Captions With Deep Learning Yulia Kogan & Ron Shiff
A First Look at Music Composition using LSTM Recurrent Neural Networks
Final Presentation: Neural Network Doc Summarization
RNNs & LSTM Hadar Gorodissky Niv Haim.
Understanding LSTM Networks
Artificial Neural Networks
ECE599/692 - Deep Learning Lecture 14 – Recurrent Neural Network (RNN)
Introduction to RNNs for NLP
Code Completion with Neural Attention and Pointer Networks
The use of Neural Networks to schedule flow-shop with dynamic job arrival ‘A Multi-Neural Network Learning for lot Sizing and Sequencing on a Flow-Shop’
Long Short Term Memory within Recurrent Neural Networks
Neural Networks Geoff Hulten.
Other Classification Models: Recurrent Neural Network (RNN)
Department of Electrical Engineering
Lecture 16: Recurrent Neural Networks (RNNs)
RNNs: Going Beyond the SRN in Language Prediction
Improving Collimator Setup Efficiency
LSTM: Long Short Term Memory
Deep Learning for the Soft Cutoff Problem
Meta Learning (Part 2): Gradient Descent as LSTM
Attention for translation
Kostas Kolomvatsos, Christos Anagnostopoulos
Automatic Handwriting Generation
The Updated experiment based on LSTM
Introduction to Neural Networks
Modeling IDS using hybrid intelligent systems
Neural Machine Translation using CNN
Recurrent Neural Networks
Deep learning: Recurrent Neural Networks CV192
CRCV REU 2019 Kara Schatz.
Week 7 Presentation Ngoc Ta Aidean Sharghi
Operational Results of LHC Collimator Alignment using Machine Learning
Beating the market -- forecasting the S&P 500 Index
Neural Machine Translation by Jointly Learning to Align and Translate
Operational Results of LHC Collimator Alignment using Machine Learning
Presentation transcript:

LHC beam mode classification Dr Ing Gianluca Valentino Department of Communications and Computer Engineering University of Malta Data Science Research Group

Outline Introduction – LHC Machine Cycle Features used LSTM model training Results Conclusions

Introduction – LHC Machine Cycle Adjust Injection Ramp Flat Top Stable Beams Squeeze

Problem formulation We want to build a classifier which can predict transitions amongst the following four beam modes using only beam loss data: Transition between injection to ramp Transition between ramp and flat top Transition between flat top and squeeze Transition between squeeze and collisions

Dataset generation There are ~3600 Beam Loss Monitors (BLM) around the LHC to measure local beam losses. They provide 1 Hz data at various running sums (RS09 = 1.31 s chosen, similar to what is normally used in multi-turn loss analysis. Data obtained from Timber (CERN logging service), from 168 pp physics fills. +/- 50 seconds around change in beam mode. Considered only losses from the 42 BLMs at IR7 collimators. Therefore each training sample has dimension 100 x 42. BLM Ionization Chambers in LHC

Start Squeeze Start Flat Top Some examples of the multivariate time-series BLM data Start Ramp Start Adjust

Dataset generation Feature scaling: Summary of classes: Beam mode transition # samples in dataset Start of ramp 168 Start of flat top 166 Start of squeeze 131 Start of adjust 151 Total 616 Feature scaling: Each multivariate time series was normalized by the BLM signal at the TCP.C6L7.B1 (primary) collimator - where we generally expect highest losses.

Recurrent Neural Network Motivation: Not all problems can be solved with a neural network structure having a fixed number of inputs and outputs. Measurement data is often sequential (time-series) In practice we can have different input/output scenarios

Recurrent Neural Network Output Sequence Output Yt Yt-1 Yt Yt+1 wy wy wh wh … … = ht ht-1 ht ht+1 wx wx Xt Xt-1 Xt Xt+1 Input Input Sequence

Problem of long-term dependencies Consider the difference between: “The clouds are in the sky”. “I lived in France for three years while I was working for a software development company. I can speak fluent French”. Classical RNNs are not capable of learning these long-term dependencies -> we need LSTMs.

Long Short Term Memory (LSTM) Instead of a single neural network, there are four networks.

LSTM cell state The cell state is the horizontal line running across the top Information can flow along it unchanged or with minor modifications.

“Forget gate” layer Decides which values of cell to reset

“Input gate” layer Sigmoid layer: decides which values of cell to write to Tanh layer: creates vector of new candidate values to write to cell

Update cell state The LSTM applies the decisions to the memory cell:

“Output gate” layer A sigmoid layer decides which values of cell to output.

ML training A RNN-LSTM model was trained to predict the output class. The output of the LSTM was forwarded to a single Dense layer of size 4, each with a softmax activation function One-hot encoding was used to represent the output (e.g. class #2 -> [0,0,1,0]) Train/test ratio used: 80% / 20%. Implementation: Python and Keras.

ML training Cross-validation was done to determine the best parameters for the following: Number of LSTM neurons: [8, 16, 32, 64, 128, 256] -> 32 picked as best Optimizer: [Adam, SGD, RMSprop, Adadelta] -> Adam picked as best Dropout: [0.1, 0.2, 0.3, 0.4] -> 0.2 picked as best

Results Accuracy and loss on training and testing sets

Results Classification Report from Scikit-Learn: Beam mode transition precision recall f1-score Start of ramp 1.00 0.94 0.97 Start of flat top 0.83 0.89 Start of squeeze 0.75 0.86 0.80 Start of adjust 0.72 0.84 0.78

Conclusions Demonstrated applicability of LSTMs in classifying transitions in beam mode First time (I believe) RNNs used on LHC data Achieved accuracy on unseen data of ~87% Future work: try also using orbit data from BPMs