Recurrent neural networks Modelling sequential data. MLP Lecture 9 Recurrent Networks 1
|
|
- Lynette Gray
- 6 years ago
- Views:
Transcription
1 Recurrent neural networks Modelling sequential data MLP Lecture 9 Recurrent Networks 1
2 Recurrent Networks Steve Renals Machine Learning Practical MLP Lecture 9 16 November 2016 MLP Lecture 9 Recurrent Networks 2
3 Introduction - Recurrent Neural Networks (RNNs) Modelling sequential data Recurrent hidden unit connections Training RNNs: Back-propagation through time LSTMs Examples (speech and language) MLP Lecture 9 Recurrent Networks 3
4 Sequential Data output Modelling sequential data with time dependences between feature vectors hidden input x1 x2 x3 x1 x2 x3 x1 x2 x3 t-2 t-1 t 2 frames of context MLP Lecture 9 Recurrent Networks 4
5 Sequential Data input output hidden x1 x2 x3 x1 x2 x3 t-2 t-1 2 frames of context x1 x2 x3 t Modelling sequential data with time dependences between feature vectors Can model fixed context with a feed-forward network with previous time input vectors added to the network input Finite context determined by window width MLP Lecture 9 Recurrent Networks 4
6 Sequential Data recurrent hidden output input x1 x2 x3 t Modelling sequential data with time dependences between feature vectors Can model fixed context with a feed-forward network with previous time input vectors added to the network input Finite context determined by window width Model sequential inputs using recurrent connections to learn a time-dependent state Potentially infinite context MLP Lecture 9 Recurrent Networks 4
7 Recurrent networks If there was no external input... think of recurrent networks in terms of the dynamics of the recurrent hidden state Settle to a fixed point stable representation Regular oscillation ( limit cycle ) learn some kind of repetition Chaotic dynamics (non-repetitive) theoretically interesting ( computation at the edge of chaos ) Useful behaviours of recurrent networks with external inputs: Recurrent state as memory remember things for (potentially) an infinite time Recurrent state as information compression compress a sequence into a state representation MLP Lecture 9 Recurrent Networks 5
8 Vanilla RNNs MLP Lecture 9 Recurrent Networks 6
9 Simplest recurrent network y k (t) = softmax ( H r=0 d h j (t) = sigmoid s=0 w (2) kr h r (t) b k w (1) js x s (t) H r=0 ) jr h r (t 1) b j w (R) } {{ } Recurrent part Output (t) w (2) Hidden (t) w (1) w (R) Input (t) Hidden (t-1) MLP Lecture 9 Recurrent Networks 7
10 Recurrent network unfolded in time Output (t-1) Output (t) Output (t1) w (2) w (2) w (2) w (R) Hidden (t-1) w (R) Hidden (t) w (R) Hidden (t1) w (R) w (1) w (1) w (1) Input (t-1) Input (t) Input (t1) An RNN for a sequence of T inputs can be viewed as a deep T -layer network with shared weights MLP Lecture 9 Recurrent Networks 8
11 Recurrent network unfolded in time Output (t-1) Output (t) Output (t1) w (2) w (2) w (2) w (R) Hidden (t-1) w (R) Hidden (t) w (R) Hidden (t1) w (R) w (1) w (1) w (1) Input (t-1) Input (t) Input (t1) An RNN for a sequence of T inputs can be viewed as a deep T -layer network with shared weights MLP Lecture 9 Recurrent Networks 8
12 Recurrent network unfolded in time Output (t-1) Output (t) Output (t1) w (2) w (2) w (2) w (R) Hidden (t-1) w (R) Hidden (t) w (R) Hidden (t1) w (R) w (1) w (1) w (1) Input (t-1) Input (t) Input (t1) An RNN for a sequence of T inputs can be viewed as a deep T -layer network with shared weights We can train an RNN by doing backprop through this unfolded network, making sure we share the weights Weight sharing if two weights are constrained to be equal (w 1 = w 2 ) then they will stay equal if the weight changes are equal ( E/ w 1 = E/ w 2 ) achieve this by updating with ( E/ w 1 E/ w 2 ) (cf Conv Nets) MLP Lecture 9 Recurrent Networks 8
13 Back-propagation through time (BPTT) We can train a network by unfolding and back-propagating through time, summing the derivatives for each weight as we go through the sequence More efficiently, run as a recurrent network cache the unit outputs at each timestep cache the output errors at each timestep then backprop from the final timestep to zero, computing the derivatives at each step compute the weight updates by summing the derivatives across time Expensive backprop for a 1,000 item sequence equivalent to a 1,000-layer feed-forward network Truncated BPTT backprop through just a few time steps (e.g. 20) MLP Lecture 9 Recurrent Networks 9
14 Vanishing and exploding gradients BPTT involves taking the product of many gradients (as in a very deep network) this can lead to vanishing (component gradients less than 1) or exploding (greater than 1) gradients This can prevent effective training Modified optimisation algorithms RMSProp (and similar algorithms) normalise the gradient for each weight by average of it magnitude, with a learning rate for each weight Hessian-free an approximation to second-order approaches which use curvature information Modified hidden unit transfer functions Long short term memory (LSTM) Linear self-recurrence for each hidden unit (long-term memory) Gates - dynamic weights which are a function of their inputs Gated recurrent units MLP Lecture 9 Recurrent Networks 10
15 LSTM MLP Lecture 9 Recurrent Networks 11
16 Vanilla RNN h(t) g(t) Whh Whx x(t) g(t) = W hx x(t) W hh h(t 1) b h h(t) = tanh (g(t)) MLP Lecture 9 Recurrent Networks 12
17 LSTM Internal recurrent state ( cell ) c(t) combines previous state c(t 1) and LSTM input g(t) MLP Lecture 9 Recurrent Networks 13
18 LSTM h(t) g(t) W hh W hx x(t) MLP Lecture 9 Recurrent Networks 14
19 LSTM Internal recurrent state h(t) c(t-1) c(t) g(t) W hh W hx x(t) MLP Lecture 9 Recurrent Networks 14
20 LSTM Internal recurrent state ( cell ) c(t) combines previous state c(t 1) and LSTM input g(t) Gates - weights dependent on the current input and the previous state Input gate: controls how much input to the unit g(t) is written to the internal state c(t) Forget gate: controls how much of the previous internal state c(t 1) is written to the internal state c(t) Input and forget gates together allow the network to control what information is stored and overwritten at each step MLP Lecture 9 Recurrent Networks 15
21 LSTM h(t) c(t-1) c(t) g(t) W hh W hx x(t) MLP Lecture 9 Recurrent Networks 16
22 LSTM Input Gate h(t) c(t-1) c(t) I(t; x(t), ) g(t) W hh W hx x(t) MLP Lecture 9 Recurrent Networks 16
23 LSTM Forget Gate h(t) c(t-1) F(t; x(t), ) c(t) I(t; x(t), ) g(t) W hh W hx x(t) MLP Lecture 9 Recurrent Networks 16
24 LSTM Input and Forget Gates h(t) c(t-1) F(t; x(t), ) c(t) I(t; x(t), ) g(t) Whh Whx I(t) = σ (W ix x(t) W ih h(t 1) b i ) F(t) = σ (W fx x(t) W fh h(t 1) b f ) σ is the sigmoid function x(t) g(t) = W hx x(t) W hh h(t 1) b h c(t) = F(t) c(t 1) I(t) g(t) is element-wise vector multiply MLP Lecture 9 Recurrent Networks 17
25 LSTM Internal recurrent state ( cell ) c(t) combines previous state c(t 1) and LSTM input g(t) Gates - weights dependent on the current input and the previous state Input gate: controls how much input to the unit g(t) is written to the internal state c(t) Forget gate: controls how much of the previous internal state c(t 1) is written to the internal state c(t) Input and forget gates together allow the network to control what information is stored and overwritten at each step Output gate: controls how much of each unit s activation is output by the hidden state it allows the LSTM cell to kepp information that is not relevant at the current time, but may be relevant later MLP Lecture 9 Recurrent Networks 18
26 LSTM Input and Forget Gates h(t) c(t-1) F(t; x(t), ) c(t) I(t; x(t), ) g(t) W hh W hx x(t) MLP Lecture 9 Recurrent Networks 19
27 LSTM Output Gate h(t) c(t-1) F(t; x(t), ) O(t; x(t), ) c(t) I(t; x(t), ) g(t) W hh W hx x(t) MLP Lecture 9 Recurrent Networks 19
28 LSTM Output Gate h(t) c(t-1) F(t; x(t), ) O(t; x(t), ) c(t) I(t; x(t), ) g(t) Whh Whx O(t) = σ (W oxx(t) W oh h(t 1) b o) x(t) h(t) = tanh (O(t) c(t)) MLP Lecture 9 Recurrent Networks 20
29 LSTM h(t) c(t-1) F(t; x(t), ) O(t; x(t), ) c(t) I(t; x(t), ) g(t) Whh Whx x(t) I(t) = σ (W ix x(t) W ih h(t 1) b i ) F(t) = σ (W fx x(t) W fh ht 1) b f ) O(t) = σ (W oxx(t) W oh h(t 1) b o) g(t) = W hx x(t) W hh h(t 1) b h c(t) = F(t) c(t 1) I(t) g(t) h(t) = tanh (O(t) c(t)) MLP Lecture 9 Recurrent Networks 21
30 Example applications using RNNs MLP Lecture 9 Recurrent Networks 22
31 Example 1: speech recognition with recurrent networks Phoneme Probabilities Recurrent Neural Network freq (Hz) Speech Acoustics time (ms) T Robinson et al (1996). The use of recurrent networks in continuous speech recognition, in Automatic Speech and Speaker Recognition Advanced Topics (Lee et al (eds)), Kluwer, MLP Lecture 9 Recurrent Networks 23
32 Example 2: speech recognition with stacked LSTMs input input input input LSTM LSTM LSTM LSTM output (a) LSTM LSTM output recurrent output (c) LSTMP recurrent LSTM (b) DLSTM recurrent output (d) DLSTMP H Sak et al (2014). Long Short-Term Memory based Recurrent Neural Network Architectures for Large Scale Acoustic Modelling, Interspeech. MLP Lecture 9 Recurrent Networks 24
33 Example 3: recurrent network language models T Mikolov et al (2010). Recurrent Neural Network Based Language Model, Interspeech MLP Lecture 9 Recurrent Networks 25
34 Example 4: recurrent encoder-decoder Machine translation I Sutskever et al (2014). Sequence to Sequence Learning with Neural Networks, NIPS. K Cho et al (2014). Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation, EMNLP. MLP Lecture 9 Recurrent Networks 26
35 Summary RNNs can model sequences Unfolding an RNN gives a deep feed-forward network Back-propagation through time LSTM More on recurrent networks next semester in NLU (and 1-2 lectures in ASR and MT) MLP Lecture 9 Recurrent Networks 27
Recurrent neural networks Modelling sequential data. MLP Lecture 9 / 13 November 2018 Recurrent Neural Networks 1: Modelling sequential data 1
Recurrent neural networks Modelling sequential data MLP Lecture 9 / 13 November 2018 Recurrent Neural Networks 1: Modelling sequential data 1 Recurrent Neural Networks 1: Modelling sequential data Steve
More informationRecurrent neural networks Modelling sequential data. MLP Lecture 9 Recurrent Neural Networks 1: Modelling sequential data 1
Recurrent neural networks Modelling sequential data MLP Lecture 9 Recurrent Neural Networks 1: Modelling sequential data 1 Recurrent Neural Networks 1: Modelling sequential data Steve Renals Machine Learning
More informationNeural Network Part 4: Recurrent Neural Networks
Neural Network Part 4: Recurrent Neural Networks Yingyu Liang Computer Sciences 760 Fall 2017 http://pages.cs.wisc.edu/~yliang/cs760/ Some of the slides in these lectures have been adapted/borrowed from
More informationDeep Learning Basics Lecture 9: Recurrent Neural Networks. Princeton University COS 495 Instructor: Yingyu Liang
Deep Learning Basics Lecture 9: Recurrent Neural Networks Princeton University COS 495 Instructor: Yingyu Liang Introduction Recurrent neural networks Dates back to (Rumelhart et al., 1986) A family of
More informationRobustness (cont.); End-to-end systems
Robustness (cont.); End-to-end systems Steve Renals Automatic Speech Recognition ASR Lecture 18 27 March 2017 ASR Lecture 18 Robustness (cont.); End-to-end systems 1 Robust Speech Recognition ASR Lecture
More information11/13/18. Introduction to RNNs for NLP. About Me. Overview SHANG GAO
Introduction to RNNs for NLP SHANG GAO About Me PhD student in the Data Science and Engineering program Took Deep Learning last year Work in the Biomedical Sciences, Engineering, and Computing group at
More informationAre there alternatives to Sigmoid Hidden Units? MLP Lecture 6 Hidden Units / Initialisation 1
Are there alternatives to Sigmoid Hidden Units? MLP Lecture 6 Hidden Units / Initialisation 1 Hidden Unit Transfer Functions Initialising Deep Networks Steve Renals Machine Learning Practical MLP Lecture
More informationGenerating an appropriate sound for a video using WaveNet.
Australian National University College of Engineering and Computer Science Master of Computing Generating an appropriate sound for a video using WaveNet. COMP 8715 Individual Computing Project Taku Ueki
More informationCoursework 2. MLP Lecture 7 Convolutional Networks 1
Coursework 2 MLP Lecture 7 Convolutional Networks 1 Coursework 2 - Overview and Objectives Overview: Use a selection of the techniques covered in the course so far to train accurate multi-layer networks
More information신경망기반자동번역기술. Konkuk University Computational Intelligence Lab. 김강일
신경망기반자동번역기술 Konkuk University Computational Intelligence Lab. http://ci.konkuk.ac.kr kikim01@kunkuk.ac.kr 김강일 Index Issues in AI and Deep Learning Overview of Machine Translation Advanced Techniques in
More informationarxiv: v1 [cs.ne] 5 Feb 2014
LONG SHORT-TERM MEMORY BASED RECURRENT NEURAL NETWORK ARCHITECTURES FOR LARGE VOCABULARY SPEECH RECOGNITION Haşim Sak, Andrew Senior, Françoise Beaufays Google {hasim,andrewsenior,fsb@google.com} arxiv:12.1128v1
More informationDeep Neural Networks (2) Tanh & ReLU layers; Generalisation and Regularisation
Deep Neural Networks (2) Tanh & ReLU layers; Generalisation and Regularisation Steve Renals Machine Learning Practical MLP Lecture 4 9 October 2018 MLP Lecture 4 / 9 October 2018 Deep Neural Networks (2)
More informationArtificial Neural Networks. Artificial Intelligence Santa Clara, 2016
Artificial Neural Networks Artificial Intelligence Santa Clara, 2016 Simulate the functioning of the brain Can simulate actual neurons: Computational neuroscience Can introduce simplified neurons: Neural
More informationAutomatic Speech Recognition (CS753)
Automatic Speech Recognition (CS753) Lecture 9: Brief Introduction to Neural Networks Instructor: Preethi Jyothi Feb 2, 2017 Final Project Landscape Tabla bol transcription Music Genre Classification Audio
More informationAudio Effects Emulation with Neural Networks
Escola Tècnica Superior d Enginyeria Informàtica Universitat Politècnica de València Audio Effects Emulation with Neural Networks Trabajo Fin de Grado Grado en Ingeniería Informática Autor: Omar del Tejo
More informationAudio Effects Emulation with Neural Networks
DEGREE PROJECT IN TECHNOLOGY, FIRST CYCLE, 15 CREDITS STOCKHOLM, SWEDEN 2017 Audio Effects Emulation with Neural Networks OMAR DEL TEJO CATALÁ LUIS MASÍA FUSTER KTH ROYAL INSTITUTE OF TECHNOLOGY SCHOOL
More informationAttention-based Multi-Encoder-Decoder Recurrent Neural Networks
Attention-based Multi-Encoder-Decoder Recurrent Neural Networks Stephan Baier 1, Sigurd Spieckermann 2 and Volker Tresp 1,2 1- Ludwig Maximilian University Oettingenstr. 67, Munich, Germany 2- Siemens
More informationMusic Recommendation using Recurrent Neural Networks
Music Recommendation using Recurrent Neural Networks Ashustosh Choudhary * ashutoshchou@cs.umass.edu Mayank Agarwal * mayankagarwa@cs.umass.edu Abstract A large amount of information is contained in the
More informationA simple RNN-plus-highway network for statistical
ISSN 1346-5597 NII Technical Report A simple RNN-plus-highway network for statistical parametric speech synthesis Xin Wang, Shinji Takaki, Junichi Yamagishi NII-2017-003E Apr. 2017 A simple RNN-plus-highway
More informationAN IMPROVED NEURAL NETWORK-BASED DECODER SCHEME FOR SYSTEMATIC CONVOLUTIONAL CODE. A Thesis by. Andrew J. Zerngast
AN IMPROVED NEURAL NETWORK-BASED DECODER SCHEME FOR SYSTEMATIC CONVOLUTIONAL CODE A Thesis by Andrew J. Zerngast Bachelor of Science, Wichita State University, 2008 Submitted to the Department of Electrical
More informationDeep Neural Network Architectures for Modulation Classification
Deep Neural Network Architectures for Modulation Classification Xiaoyu Liu, Diyu Yang, and Aly El Gamal School of Electrical and Computer Engineering Purdue University Email: {liu1962, yang1467, elgamala}@purdue.edu
More informationAnnouncements. Today. Speech and Language. State Path Trellis. HMMs: MLE Queries. Introduction to Artificial Intelligence. V22.
Introduction to Artificial Intelligence Announcements V22.0472-001 Fall 2009 Lecture 19: Speech Recognition & Viterbi Decoding Rob Fergus Dept of Computer Science, Courant Institute, NYU Slides from John
More informationMINE 432 Industrial Automation and Robotics
MINE 432 Industrial Automation and Robotics Part 3, Lecture 5 Overview of Artificial Neural Networks A. Farzanegan (Visiting Associate Professor) Fall 2014 Norman B. Keevil Institute of Mining Engineering
More informationFEATURE COMBINATION AND STACKING OF RECURRENT AND NON-RECURRENT NEURAL NETWORKS FOR LVCSR
FEATURE COMBINATION AND STACKING OF RECURRENT AND NON-RECURRENT NEURAL NETWORKS FOR LVCSR Christian Plahl 1, Michael Kozielski 1, Ralf Schlüter 1 and Hermann Ney 1,2 1 Human Language Technology and Pattern
More informationDeep learning architectures for music audio classification: a personal (re)view
Deep learning architectures for music audio classification: a personal (re)view Jordi Pons jordipons.me @jordiponsdotme Music Technology Group Universitat Pompeu Fabra, Barcelona Acronyms MLP: multi layer
More informationNeural Turing Machines
Neural Turing Machines Can neural nets learn programs? Alex Graves Greg Wayne Ivo Danihelka Contents 1. IntroducBon 2. FoundaBonal Research 3. Neural Turing Machines 4. Experiments 5. Conclusions IntroducBon
More informationLecture 2. Digital Basics
Lecture Digital Basics Peter Cheung Department of Electrical & Electronic Engineering Imperial College London URL: www.ee.ic.ac.uk/pcheung/teaching/de1_ee/ E-mail: p.cheung@imperial.ac.uk Lecture Slide
More informationUsing Deep Learning for Sentiment Analysis and Opinion Mining
Using Deep Learning for Sentiment Analysis and Opinion Mining Gauging opinions is faster and more accurate. Abstract How does a computer analyze sentiment? How does a computer determine if a comment or
More informationEndpoint Detection using Grid Long Short-Term Memory Networks for Streaming Speech Recognition
INTERSPEECH 2017 August 20 24, 2017, Stockholm, Sweden Endpoint Detection using Grid Long Short-Term Memory Networks for Streaming Speech Recognition Shuo-Yiin Chang, Bo Li, Tara N. Sainath, Gabor Simko,
More informationAttention-based Information Fusion using Multi-Encoder-Decoder Recurrent Neural Networks
Attention-based Information Fusion using Multi-Encoder-Decoder Recurrent Neural Networks Stephan Baier1, Sigurd Spieckermann2 and Volker Tresp1,2 1- Ludwig Maximilian University Oettingenstr. 67, Munich,
More informationREAL TIME EMULATION OF PARAMETRIC GUITAR TUBE AMPLIFIER WITH LONG SHORT TERM MEMORY NEURAL NETWORK
REAL TIME EMULATION OF PARAMETRIC GUITAR TUBE AMPLIFIER WITH LONG SHORT TERM MEMORY NEURAL NETWORK Thomas Schmitz and Jean-Jacques Embrechts 1 1 Department of Electrical Engineering and Computer Science,
More informationMachine Learning in Indoor Positioning and Channel Prediction Systems. Yizhou Zhu B.Eng., Zhejiang University, 2010
Machine Learning in Indoor Positioning and Channel Prediction Systems by Yizhou Zhu B.Eng., Zhejiang University, 2010 A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of MASTER
More informationTraining neural network acoustic models on (multichannel) waveforms
View this talk on YouTube: https://youtu.be/si_8ea_ha8 Training neural network acoustic models on (multichannel) waveforms Ron Weiss in SANE 215 215-1-22 Joint work with Tara Sainath, Kevin Wilson, Andrew
More informationGated Recurrent Convolution Neural Network for OCR
Gated Recurrent Convolution Neural Network for OCR Jianfeng Wang amd Xiaolin Hu Presented by Boyoung Kim February 2, 2018 Boyoung Kim (SNU) RNN-NIPS2017 February 2, 2018 1 / 11 Optical Charactor Recognition(OCR)
More informationInvestigating Very Deep Highway Networks for Parametric Speech Synthesis
9th ISCA Speech Synthesis Workshop September, Sunnyvale, CA, USA Investigating Very Deep Networks for Parametric Speech Synthesis Xin Wang,, Shinji Takaki, Junichi Yamagishi,, National Institute of Informatics,
More informationarxiv: v2 [cs.cl] 20 Feb 2018
IMPROVED TDNNS USING DEEP KERNELS AND FREQUENCY DEPENDENT GRID-RNNS F. L. Kreyssig, C. Zhang, P. C. Woodland Cambridge University Engineering Dept., Trumpington St., Cambridge, CB2 1PZ U.K. {flk24,cz277,pcw}@eng.cam.ac.uk
More informationIntroduction to Machine Learning
Introduction to Machine Learning Perceptron Barnabás Póczos Contents History of Artificial Neural Networks Definitions: Perceptron, Multi-Layer Perceptron Perceptron algorithm 2 Short History of Artificial
More informationArtificial Intelligence and Deep Learning
Artificial Intelligence and Deep Learning Cars are now driving themselves (far from perfectly, though) Speaking to a Bot is No Longer Unusual March 2016: World Go Champion Beaten by Machine AI: The Upcoming
More informationMultiple-Layer Networks. and. Backpropagation Algorithms
Multiple-Layer Networks and Algorithms Multiple-Layer Networks and Algorithms is the generalization of the Widrow-Hoff learning rule to multiple-layer networks and nonlinear differentiable transfer functions.
More informationConvolutional neural networks
Convolutional neural networks Themes Curriculum: Ch 9.1, 9.2 and http://cs231n.github.io/convolutionalnetworks/ The simple motivation and idea How it s done Receptive field Pooling Dilated convolutions
More informationDeep Learning. Dr. Johan Hagelbäck.
Deep Learning Dr. Johan Hagelbäck johan.hagelback@lnu.se http://aiguy.org Image Classification Image classification can be a difficult task Some of the challenges we have to face are: Viewpoint variation:
More informationLearning to Unlearn and Relearn Speech Signal Processing using Neural Networks: current and future perspectives
Learning to Unlearn and Relearn Speech Signal Processing using Neural Networks: current and future perspectives Mathew Magimai Doss Collaborators: Vinayak Abrol, Selen Hande Kabil, Hannah Muckenhirn, Dimitri
More information(i) Understanding the basic concepts of signal modeling, correlation, maximum likelihood estimation, least squares and iterative numerical methods
Tools and Applications Chapter Intended Learning Outcomes: (i) Understanding the basic concepts of signal modeling, correlation, maximum likelihood estimation, least squares and iterative numerical methods
More informationDigital Integrated CircuitDesign
Digital Integrated CircuitDesign Lecture 13 Building Blocks (Multipliers) Register Adder Shift Register Adib Abrishamifar EE Department IUST Acknowledgement This lecture note has been summarized and categorized
More informationRevision of Channel Coding
Revision of Channel Coding Previous three lectures introduce basic concepts of channel coding and discuss two most widely used channel coding methods, convolutional codes and BCH codes It is vital you
More informationMachine Learning Practical Part 2: Group Projects. MLP Lecture 11 MLP Part 2: Group Projects 1
Machine Learning Practical Part 2: Group Projects MLP Lecture 11 MLP Part 2: Group Projects 1 MLP Part 2: Group Projects Steve Renals Machine Learning Practical MLP Lecture 11 24 January 2018 http://www.inf.ed.ac.uk/teaching/courses/mlp/
More informationThe Munich 2011 CHiME Challenge Contribution: BLSTM-NMF Speech Enhancement and Recognition for Reverberated Multisource Environments
The Munich 2011 CHiME Challenge Contribution: BLSTM-NMF Speech Enhancement and Recognition for Reverberated Multisource Environments Felix Weninger, Jürgen Geiger, Martin Wöllmer, Björn Schuller, Gerhard
More informationA Comparison of MLP, RNN and ESN in Determining Harmonic Contributions from Nonlinear Loads
A Comparison of MLP, RNN and ESN in Determining Harmonic Contributions from Nonlinear Loads Jing Dai, Pinjia Zhang, Joy Mazumdar, Ronald G Harley and G K Venayagamoorthy 3 School of Electrical and Computer
More informationBehaviour Patterns Evolution on Individual and Group Level. Stanislav Slušný, Roman Neruda, Petra Vidnerová. CIMMACS 07, December 14, Tenerife
Behaviour Patterns Evolution on Individual and Group Level Stanislav Slušný, Roman Neruda, Petra Vidnerová Department of Theoretical Computer Science Institute of Computer Science Academy of Science of
More informationUnsupervised Minimax: nets that fight each other
Unsupervised Minimax: nets that fight each other Jürgen Schmidhuber The Swiss AI Lab IDSIA Univ. Lugano & SUPSI http://www.idsia.ch/~juergen NNAISENSE Jürgen Schmidhuber You_again Shmidhoobuh Supervised
More informationCSC321 Lecture 11: Convolutional Networks
CSC321 Lecture 11: Convolutional Networks Roger Grosse Roger Grosse CSC321 Lecture 11: Convolutional Networks 1 / 35 Overview What makes vision hard? Vison needs to be robust to a lot of transformations
More informationContinuous time and Discrete time Signals and Systems
Continuous time and Discrete time Signals and Systems 1. Systems in Engineering A system is usually understood to be an engineering device in the field, and a mathematical representation of this system
More informationI am very pleased to teach this class again, after last year s course on electronics over the Summer Term. Based on the SOLE survey result, it is clear that the format, style and method I used worked with
More informationNeural Networks The New Moore s Law
Neural Networks The New Moore s Law Chris Rowen, PhD, FIEEE CEO Cognite Ventures December 216 Outline Moore s Law Revisited: Efficiency Drives Productivity Embedded Neural Network Product Segments Efficiency
More informationDigital Communication System
Digital Communication System Purpose: communicate information at certain rate between geographically separated locations reliably (quality) Important point: rate, quality spectral bandwidth requirement
More informationINTRODUCTION TO DEEP LEARNING. Steve Tjoa June 2013
INTRODUCTION TO DEEP LEARNING Steve Tjoa kiemyang@gmail.com June 2013 Acknowledgements http://ufldl.stanford.edu/wiki/index.php/ UFLDL_Tutorial http://youtu.be/ayzoubkuf3m http://youtu.be/zmnoatzigik 2
More informationSignal Characteristics
Data Transmission The successful transmission of data depends upon two factors:» The quality of the transmission signal» The characteristics of the transmission medium Some type of transmission medium
More informationLearning New Articulator Trajectories for a Speech Production Model using Artificial Neural Networks
Learning New Articulator Trajectories for a Speech Production Model using Artificial Neural Networks C. S. Blackburn and S. J. Young Cambridge University Engineering Department (CUED), England email: csb@eng.cam.ac.uk
More informationSpeech Coding in the Frequency Domain
Speech Coding in the Frequency Domain Speech Processing Advanced Topics Tom Bäckström Aalto University October 215 Introduction The speech production model can be used to efficiently encode speech signals.
More informationReverse Correlation for analyzing MLP Posterior Features in ASR
Reverse Correlation for analyzing MLP Posterior Features in ASR Joel Pinto, G.S.V.S. Sivaram, and Hynek Hermansky IDIAP Research Institute, Martigny École Polytechnique Fédérale de Lausanne (EPFL), Switzerland
More informationVQ Source Models: Perceptual & Phase Issues
VQ Source Models: Perceptual & Phase Issues Dan Ellis & Ron Weiss Laboratory for Recognition and Organization of Speech and Audio Dept. Electrical Eng., Columbia Univ., NY USA {dpwe,ronw}@ee.columbia.edu
More informationRadio Deep Learning Efforts Showcase Presentation
Radio Deep Learning Efforts Showcase Presentation November 2016 hume@vt.edu www.hume.vt.edu Tim O Shea Senior Research Associate Program Overview Program Objective: Rethink fundamental approaches to how
More informationImage Manipulation Detection using Convolutional Neural Network
Image Manipulation Detection using Convolutional Neural Network Dong-Hyun Kim 1 and Hae-Yeoun Lee 2,* 1 Graduate Student, 2 PhD, Professor 1,2 Department of Computer Software Engineering, Kumoh National
More informationCS 229, Project Progress Report SUNet ID: Name: Ajay Shanker Tripathi
CS 229, Project Progress Report SUNet ID: 06044535 Name: Ajay Shanker Tripathi Title: Voice Transmogrifier: Spoofing My Girlfriend s Voice Project Category: Audio and Music The project idea is an easy-to-state
More informationPerformance Analysis of MFCC and LPCC Techniques in Automatic Speech Recognition
www.ijecs.in International Journal Of Engineering And Computer Science ISSN:2319-7242 Volume - 3 Issue - 8 August, 2014 Page No. 7727-7732 Performance Analysis of MFCC and LPCC Techniques in Automatic
More informationA Technique for Pulse RADAR Detection Using RRBF Neural Network
Proceedings of the World Congress on Engineering 22 Vol II WCE 22, July 4-6, 22, London, U.K. A Technique for Pulse RADAR Detection Using RRBF Neural Network Ajit Kumar Sahoo, Ganapati Panda and Babita
More informationConstant False Alarm Rate Detection of Radar Signals with Artificial Neural Networks
Högskolan i Skövde Department of Computer Science Constant False Alarm Rate Detection of Radar Signals with Artificial Neural Networks Mirko Kück mirko@ida.his.se Final 6 October, 1996 Submitted by Mirko
More informationLearning the Speech Front-end With Raw Waveform CLDNNs
INTERSPEECH 2015 Learning the Speech Front-end With Raw Waveform CLDNNs Tara N. Sainath, Ron J. Weiss, Andrew Senior, Kevin W. Wilson, Oriol Vinyals Google, Inc. New York, NY, U.S.A {tsainath, ronw, andrewsenior,
More informationReal-time Traffic Data Prediction with Basic Safety Messages using Kalman-Filter based Noise Reduction Model and Long Short-term Memory Neural Network
Real-time Traffic Data Prediction with Basic Safety Messages using Kalman-Filter based Noise Reduction Model and Long Short-term Memory Neural Network Mizanur Rahman*, Ph.D. Postdoctoral Fellow Center
More informationConvolutional Neural Networks for Small-footprint Keyword Spotting
INTERSPEECH 2015 Convolutional Neural Networks for Small-footprint Keyword Spotting Tara N. Sainath, Carolina Parada Google, Inc. New York, NY, U.S.A {tsainath, carolinap}@google.com Abstract We explore
More informationBiomedical Signals. Signals and Images in Medicine Dr Nabeel Anwar
Biomedical Signals Signals and Images in Medicine Dr Nabeel Anwar Noise Removal: Time Domain Techniques 1. Synchronized Averaging (covered in lecture 1) 2. Moving Average Filters (today s topic) 3. Derivative
More informationIntroduction to Machine Learning
Introduction to Machine Learning Deep Learning Barnabás Póczos Credits Many of the pictures, results, and other materials are taken from: Ruslan Salakhutdinov Joshua Bengio Geoffrey Hinton Yann LeCun 2
More informationDigital Image Processing COSC 6380/4393
Digital Image Processing COSC 638/4393 Lecture 9 Sept 26 th, 217 Pranav Mantini Slides from Dr. Shishir K Shah and Frank (Qingzhong) Liu, S. Narasimhan HISTOGRAM SHAPING We now describe methods for histogram
More informationEE228 Applications of Course Concepts. DePiero
EE228 Applications of Course Concepts DePiero Purpose Describe applications of concepts in EE228. Applications may help students recall and synthesize concepts. Also discuss: Some advanced concepts Highlight
More informationEE216B: VLSI Signal Processing. Wavelets. Prof. Dejan Marković Shortcomings of the Fourier Transform (FT)
5//0 EE6B: VLSI Signal Processing Wavelets Prof. Dejan Marković ee6b@gmail.com Shortcomings of the Fourier Transform (FT) FT gives information about the spectral content of the signal but loses all time
More information1 Introduction. w k x k (1.1)
Neural Smithing 1 Introduction Artificial neural networks are nonlinear mapping systems whose structure is loosely based on principles observed in the nervous systems of humans and animals. The major
More informationSpeech Signal Analysis
Speech Signal Analysis Hiroshi Shimodaira and Steve Renals Automatic Speech Recognition ASR Lectures 2&3 14,18 January 216 ASR Lectures 2&3 Speech Signal Analysis 1 Overview Speech Signal Analysis for
More informationLecture 23 Deep Learning: Segmentation
Lecture 23 Deep Learning: Segmentation COS 429: Computer Vision Thanks: most of these slides shamelessly adapted from Stanford CS231n: Convolutional Neural Networks for Visual Recognition Fei-Fei Li, Andrej
More informationMachine Learning for Hardware Design. Elyse Rosenbaum University of Illinois at Urbana- Champaign Oct. 18, 2017
Machine Learning for Hardware Design Elyse Rosenbaum University of Illinois at Urbana- Champaign Oct. 18, 2017 Questions, Questions, Questions 1. How can design productivity be improved? 2. What is machine
More informationFPGA-based Low-power Speech Recognition with Recurrent Neural Networks
FPGA-based Low-power Speech Recognition with Recurrent Neural Networks Minjae Lee, Kyuyeon Hwang, Jinhwan Park, Sungwook Choi, Sungho Shin and onyong Sung Department of Electrical and Computer Engineering,
More informationDigital Communication System
Digital Communication System Purpose: communicate information at required rate between geographically separated locations reliably (quality) Important point: rate, quality spectral bandwidth, power requirements
More informationSupplementary information accompanying the manuscript Biologically Inspired Modular Neural Control for a Leg-Wheel Hybrid Robot
Supplementary information accompanying the manuscript Biologically Inspired Modular Neural Control for a Leg-Wheel Hybrid Robot Poramate Manoonpong a,, Florentin Wörgötter a, Pudit Laksanacharoen b a)
More informationEur Ing Dr. Lei Zhang Faculty of Engineering and Applied Science University of Regina Canada
Eur Ing Dr. Lei Zhang Faculty of Engineering and Applied Science University of Regina Canada The Second International Conference on Neuroscience and Cognitive Brain Information BRAININFO 2017, July 22,
More informationMobile Communications TCS 455
Mobile Communications TCS 455 Dr. Prapun Suksompong prapun@siit.tu.ac.th Lecture 21 1 Office Hours: BKD 3601-7 Tuesday 14:00-16:00 Thursday 9:30-11:30 Announcements Read Chapter 9: 9.1 9.5 HW5 is posted.
More informationarxiv: v2 [cs.sd] 22 May 2017
SAMPLE-LEVEL DEEP CONVOLUTIONAL NEURAL NETWORKS FOR MUSIC AUTO-TAGGING USING RAW WAVEFORMS Jongpil Lee Jiyoung Park Keunhyoung Luke Kim Juhan Nam Korea Advanced Institute of Science and Technology (KAIST)
More informationApplication of Generalised Regression Neural Networks in Lossless Data Compression
Application of Generalised Regression Neural Networks in Lossless Data Compression R. LOGESWARAN Centre for Multimedia Communications, Faculty of Engineering, Multimedia University, 63100 Cyberjaya MALAYSIA
More informationLesson 7. Digital Signal Processors
Lesson 7 Digital Signal Processors Instructional Objectives After going through this lesson the student would learn o Architecture of a Real time Signal Processing Platform o Different Errors introduced
More informationAN ANALYSIS OF SPEECH RECOGNITION PERFORMANCE BASED UPON NETWORK LAYERS AND TRANSFER FUNCTIONS
AN ANALYSIS OF SPEECH RECOGNITION PERFORMANCE BASED UPON NETWORK LAYERS AND TRANSFER FUNCTIONS Kuldeep Kumar 1, R. K. Aggarwal 1 and Ankita Jain 2 1 Department of Computer Engineering, National Institute
More informationLesson 08. Convolutional Neural Network. Ing. Marek Hrúz, Ph.D. Katedra Kybernetiky Fakulta aplikovaných věd Západočeská univerzita v Plzni.
Lesson 08 Convolutional Neural Network Ing. Marek Hrúz, Ph.D. Katedra Kybernetiky Fakulta aplikovaných věd Západočeská univerzita v Plzni Lesson 08 Convolution we will consider 2D convolution the result
More informationPrediction by a Hybrid of Wavelet Transform and Long-Short-Term-Memory Neural Network
Prediction by a Hybrid of Wavelet Transform and Long-Short-Term-Memory Neural Network Putu Sugiartawan, Reza Pulungan, and Anny Kartika Sari Department of Computer Science and Electronics Faculty of Mathematics
More informationConvolutional Networks Overview
Convolutional Networks Overview Sargur Srihari 1 Topics Limitations of Conventional Neural Networks The convolution operation Convolutional Networks Pooling Convolutional Network Architecture Advantages
More informationAccelerating Stochastic Random Projection Neural Networks
Rochester Institute of Technology RIT Scholar Works Theses Thesis/Dissertation Collections 12-2017 Accelerating Stochastic Random Projection Neural Networks Swathika Ramakrishnan sxr1661@rit.edu Follow
More information2 TD-MoM ANALYSIS OF SYMMETRIC WIRE DIPOLE
Design of Microwave Antennas: Neural Network Approach to Time Domain Modeling of V-Dipole Z. Lukes Z. Raida Dept. of Radio Electronics, Brno University of Technology, Purkynova 118, 612 00 Brno, Czech
More informationA.I. and Translation. iflytek Research : Gao Jianqing
A.I. and Translation iflytek Research : Gao Jianqing 11-2017 1. Introduction of iflytek and A.I. 2. Application of A.I. in Translation Company Overview Founded in 1999 A leading IT Enterprise in China
More informationIntroduction (concepts and definitions)
Objectives: Introduction (digital system design concepts and definitions). Advantages and drawbacks of digital techniques compared with analog. Digital Abstraction. Synchronous and Asynchronous Systems.
More informationHOW DO DEEP CONVOLUTIONAL NEURAL NETWORKS
Under review as a conference paper at ICLR 28 HOW DO DEEP CONVOLUTIONAL NEURAL NETWORKS LEARN FROM RAW AUDIO WAVEFORMS? Anonymous authors Paper under double-blind review ABSTRACT Prior work on speech and
More informationIntroduction to Source Coding
Comm. 52: Communication Theory Lecture 7 Introduction to Source Coding - Requirements of source codes - Huffman Code Length Fixed Length Variable Length Source Code Properties Uniquely Decodable allow
More informationFixed- Weight Controller for Multiple Systems
Fixed Weight Controller for Multiple Systems L. A. Feldkamp and G. V. Puskorius Ford Research Laboratory, P.O. Box 253, MD 117 SRL Dearborn, Michigan 48 12 1253 IfeldkamQford. com, gpuskori @ford. com
More informationA Hybrid Architecture using Cross Correlation and Recurrent Neural Networks for Acoustic Tracking in Robots
A Hybrid Architecture using Cross Correlation and Recurrent Neural Networks for Acoustic Tracking in Robots John C. Murray, Harry Erwin and Stefan Wermter Hybrid Intelligent Systems School for Computing
More informationExperiments with Noise Reduction Neural Networks for Robust Speech Recognition
Experiments with Noise Reduction Neural Networks for Robust Speech Recognition Michael Trompf TR-92-035, May 1992 International Computer Science Institute, 1947 Center Street, Berkeley, CA 94704 SEL ALCATEL,
More information