Lecture Calendar

Date Room Topic References Additional Material
1 19/02/2018
(14-16)
C1 Introduction to the course: motivations and aim; course housekeeping (exams, timetable, materials); introduction to modern pattern recognition applications





 2 20/02/2018
(14-16)
 C1 Signal processing: timeseries; time domain analysis (statistics, correlation); spectral analysis (fourier analysis; wavelets).
  Demo
Matlab temperature seasonality demo
 3  26/02/2018
(14-16)
 C1 Image processing: feature descriptors (color histograms, SIFT), spectral analysis, feature detectors (edge, blobs and segments).
  Additional Readings
[1,2] Two high-level surveys on visual feature extraction and representation
 4  27/02/2018
(14-16)
 C1 Generative and Graphical Models: probability refresher; graphical model representation; directed and undirected models: Bayesian networks; Markov networks; conditional independence and d-separation.
[BRML] Ch. 1 and 2 (Refresher)
[BRML] Sect. 3.1, 3.2 and 3.3.1
(conditional independence)
[BRML] Sect. 3.3 (Directed Models)
[BRML] Sect. 4.1, 4.2.0-4.2.2 (Undirected Models)
[BRML] Sect. 4.5 (Expressiveness)
Sofware
  • Pyro - Python library based on PyTorch
  • PyMC3 - Python library based on Theano
  • Edward - Python library based on TensorFlow
 5 06/03/2018
(14-16)
 C1 Hidden Markov Models:  learning in directed graphical models; forward-backward algorithm;  generative models for sequential data
[BRML] Sect. 23.1.0 (Markov Models)
[BRML] Sect. 23.2.0-23.2.4 (HMM and forward backward) 
Additional Readings
[3]  A classical tutorial introduction to HMMs
 6 12/03/2018
(14-16)
 C1 Hidden Markov Models: part 2
   Sofware
  • hmmlearn - Scikit-learn like library for HMMs
  • hmms 0.1 - scalable discrete and continuous time HMMs in Python

 7 13/03/2018
(14-16)
 C1 Markov Random Fields: learning in undirected graphical  models; conditional random fields; pattern recognition applications
[BRML] Sect. 4.2.2, 4.2.5 (MRF)
[BRML] Sect. 4.4 (Factor Graphs)
[BRML] Sect. 5.1.1 (Variable Elimination and Inference on Chain) 
[BRML] Sect. 9.6.0, 9.6.1, 9.6.4, 9.6.5 (Learning in MRF/CRF)
Additional Readings
[4,5] Two comprehensive tutorials on CRF ([4] more introductory and [5] more focused on vision)
[6] A nice application of CRF to image segmentation

Sofware
Check out pgmpy: it has Python notebooks to introduce to working with MRF/CRF
8 19/03/2018
(14-16)
C1
Boltzmann Machines: bridging neural networks and generative models; stochastic neuron; restricted Boltzmann machine; contrastive divergence
[DL] Sections 20.1 and 20.2
Additional Readings
[7] A clean and clear introduction to RBM from its author

Sofware
Matlab code for Deep Belief Networks (i.e. stacked RBM) and Deep Boltzmann Machines.
 9 20/03/2018
(14-16)
C1
Bayesian Learning: non-parametric models; variational learning; sampling
[BRML] Sect. 11.2.1 (Variational EM), 20.4-20.6.1  (LDA)
Additional Readings
[8] LDA foundation paper
[9] A gentle introduction to latent topic models
[10] Foundations of bag of words image representation

Sofware
 M1  22/03/2018
(11-13)
 N Midterm 1 discussion
   
10 26/03/2018
(14-16)
 C1 Generative models for structures
(guest seminar by Daniele Castellana)
  Additional Readings
[11] HMM extension to bottom-up trees
[12] Mixture of trees paper

Sofware

Official Python implementation for the bottom-up tree model
 11 27/03/2018
(14-16)
C1
Deep Autoencoders: introduction to the deep learning module, sparse, denoising and contractive AE; deep RBM.
 [DL] Chapter 14
Additional Readings
[13] DBN: the paper that started deep learning
[14] Deep Boltzmann machines paper
[15] Review paper on deep generative models
[16] Long review paper on autoencoders from the perspective of representation learning
[17] Paper discussing regularized autoencoder as approximations of likelihood gradient

       Lecture break period due to Easter holidays and half-semester exams   
(28/04/2018-13/04/2018) 

   Date Room
Topic
References
Additional Material
12  16/04/2018
(14-16)
C1 Deep Autoencoders (part II) + Convolutional Neural Networks (part I):
deep RBM; introduction to CNN; basic CNN elements
[DL] Chapter 9 Additional Readings
[18-22] Original papers for LeNet, AlexNet, VGGNet, GoogLeNet and ResNet.
13 17/04/2018
C1
Convolutional Neural Networks (part II):
CNN architectures for image recognition; convolution visualization; advanced topics (deconvolution, dense nets); applications and code
[DL] Chapter 9
Additional Readings
[23] Complete summary of convolution arithmetics
[24] Seminal paper on batch normalization
[25] CNN interpretation using deconvolutions
14 19/04/2018
(11-13)
N Learning in Structured Domain I: Recursive Neural Networks
(Guest lecture by Alessio Micheli)


15 23/04/2018
(14-16)
 C1 Reservoir Computing Approaches
(Guest lecture by Claudio Gallicchio)

   
16 24/04/2018
(14-16)
 C1 Learning in Structured Domain II: Neural Networks for Graphs
(Guest lecture by Alessio Micheli)
   
 17 30/04/2018
(14-16)
 C1 Gated Recurrent Networks: deep learning for sequence processing; gradient issues; long-short term memory; gated recurrent units; generative use of RNN
[DL] Sections 10.1-10.3, 10.5-10.7, 10.10, 10.11
 Additional Readings
[26] Paper describing gradient vanish/explosion
[27] Original LSTM paper
[28] An historical view on gated RNN
[29] Gated recurren units paper
[30] Seminal paper on dropout regularization
[31] Tree LSTM for sentiment analysis using classical recursive encoding

Sofware
 M2 03/05/2018 (11-13)
 N Midterm 2 discussion    

 07/05/2018 (14-16)  C1 Lecture cancelled due to students' assembly
   

08/05/2018 (14-16)  C1 Lecture cancelled due to students' elections
   
  18 10/05/2018 (11-13)  B From TensorFlow to Keras:
python; tensorflow; keras; CNN

(Coding practice by Francesco Crecchi)
  Software
Coding examples from the lecture available on the group's GitLab
  19 14/05/2018 (14-16)  C1 PyTorch – neural networks in Python:
python; pytorch
(Coding practice by Antonio Carta)
   
 20 15/05/2018 (14-16)  C1 Advanced Recurrent Architectures: sequence-to-sequence;  attention models; multiscale network; memory networks; neural reasoning.
 [DL] Sections 10.12, 12.4.5
Additional Readings
[32,33] Models of sequence-to-sequence and image-to-sequence transduction with attention
[34,35] Models optimizing dynamic memory usage (clockwork RNN, zoneout)
[36] Differentiable memory networks
[37,38] Neural Turing Machines and follow-up paper on pondering networks
 
 21 21/05/2018 (14-16)  C1 Generative and Unsupervised Deep Learning: explicit distribution models; variational autoencoders; adversarial learning.
 [DL] Sections 20.9, 20.10.1-20.10.4
Additional Readings
[39] PixelCNN - Explict likelihood model
[40] Tutorial on VAE
[41] Tutorial on GAN (here another online resource with GAN tips)
[42] A recent stabler extension to GAN
[43] Tutorial on sampling neural networks

Sofware
 22 22/05/2018 (14-16)  C1 Solving 'omics problems with Deep Learning
(guest seminar by Marco Podda)
   
 23 24/05/2018 (11-13)
 B Fooling Deep Nets for fun and profit: On the state of Adversarial Machine Learning
(guest seminar by Francesco Crecchi)
   
 24 28/05/2018 (14-16)  C1 Memory Enhanced Networks:
attention, Neural Turing Machine, Linear Memory Network
(Guest seminar by Antonio Carta)
   
 25 29/05/2018 (14-16)  C1 Final lecture: course wrap-up; research themes; final projects; exam modalities
   
 M3 01/06/2018 (11-13)  N1 Midterm 3 discussion

Bibliography

  1. Scott Krigg, Interest Point Detector and Feature Descriptor Survey, Computer Vision Metrics, pp 217-282, Open Access Chapter
  2. Tinne Tuytelaars and Krystian Mikolajczyk, Local Invariant Feature Detectors: A Survey, Foundations and Trends in Computer Graphics and Vision, Vol. 3, No. 3 (2007) 177–2, Online Version
  3. Lawrence R. Rabiner:a tutorial on hidden Markov models and selected applications in speech recognition. Proceedings of the IEEE, 1989, pages 257-286, Online Version
  4. Charles Sutton and Andrew McCallum,  An Introduction to Conditional Random Fields, Arxiv
  5. Sebastian Nowozin and Christoph H. Lampert, Structured Learning and Prediction, Foundations and Trends in Computer Graphics and Vision, Online Version
  6. Philipp Krahenbuhl, Vladlen Koltun, Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials, Proc.of NIPS 2011, Arxiv
  7. Geoffrey Hinton, A Practical Guide to Training Restricted Boltzmann Machines, Technical Report 2010-003, University of Toronto, 2010
  8. D. Blei, A. Y. Ng, M. I. Jordan. Latent Dirichlet Allocation. Journal of Machine Learning Research, 2003
  9. D. Blei. Probabilistic topic models. Communications of the ACM, 55(4):77–84, 2012, Free Online Version
  10. G. Csurka, C. R. Dance, L. Fan, J. Willamowski, and C. Bray. Visual Categorization with Bags of Keypoints. Workshop on Statistical Learning in Computer Vision. ECCV 2004, Free Online Version
  11. D. Bacciu, A. Micheli, A. Sperduti, Compositional Generative Mapping for Tree-Structured Data - Part I: Bottom-Up Probabilistic Modeling of Trees, IEEE Transactions on Neural Networks and Learning Systems, vol. 23, no. 12, pp. 1987-2002, 2012, TR with full derivation details
  12. D. Bacciu, D. Castellana, Mixture of Hidden Markov Models as Tree Encoder,  ESANN 2018, Paper site.
  13. G.E. Hinton, R. R. Salakhutdinov. Reducing the dimensionality of data with neural networks.Science 313.5786 (2006): 504-507, Free Online Version
  14. G.E. Hinton, R. R. Salakhutdinov. Deep Boltzmann Machines. AISTATS 2009, Free online version.
  15. R. R. Salakhutdinov. Learning Deep Generative Models, Annual Review of Statistics and Its Application, 2015, Free Online Version
  16. Y. Bengio, A. Courville, and P. Vincent. Representation learning: A review and new perspectives. Pattern Analysis and Machine Intelligence, IEEE Transactions on, Vol. 35(8) (2013): 1798-1828, Arxiv.
  17. G. Alain, Y. Bengio. What Regularized Auto-Encoders Learn from the Data-Generating Distribution, JMLR, 2014.
  18. Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard and L. D. Jackel. Handwritten digit recognition with a back-propagation network, Advances in Neural Information Processing Systems, NIPS, 1989
  19. A. Krizhevsky, I. Sutskever, and G. E. Hinton. ImageNet classification with deep convolutional neural networks, Advances in Neural Information Processing Systems, NIPS, 2012
  20. S. Simonyan and A. Zisserman.  Very deep convolutional networks for large-scale image recognition, ICLR 2015, Free Online Version
  21. C. Szegedy et al,  Going Deeper with Convolutions, CVPR 2015, Free Online Version
  22. K. He, X. Zhang, S. Ren, and J. Sun. Deep Residual Learning for Image Recognition. CVPR 2016, Free Online Version
  23. V. Dumoulin, F. Visin, A guide to convolution arithmetic for deep learning, Arxiv
  24. S. Ioffe, C. Szegedy, Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift, ICML 2013,  Arxiv
  25. M.D. Zeiler and R. Fergus, Visualizing and Understanding Convolutional Networks, ICML 2013, Arxiv
  26. Y. Bengio, P. Simard and P. Frasconi, Learning long-term dependencies with gradient descent is difficult. TNN, 1994, Free Online Version
  27. S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Computation, 1997, Free Online Version
  28. K. Greff et al, LSTM: A Search Space Odyssey, TNNLS 2016, Arxiv
  29. C. Kyunghyun et al, Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation, EMNLP 2014, Arxiv
  30. N. Srivastava et al, Dropout: A Simple Way to Prevent Neural Networks from Overfitting, JLMR 2014
  31. K. Sheng et al, Improved Semantic Representations From. Tree-Structured Long Short-Term Memory Networks, ACL 2015, Arxiv
  32. Bahdanau et al, Neural machine translation by jointly learning to align and translate, ICLR 2015, Arxiv
  33. Xu et al, Show, Attend and Tell: Neural Image Caption Generation with Visual Attention, ICML 2015, Arxiv
  34. Koutník et al, A Clockwork RNN, ICML 2014, Arxiv
  35. Krueger, Zoneout: Regularizing RNNs by Randomly Preserving Hidden Activation, ICLR 2018, Arxiv
  36. Sukhbaatar et al, End-to-end Memory Networks, NIPS 2015, Arxiv
  37. Graves et al, Neural Turing Machines, Arxiv
  38. A.Graves, Adaptive Computation Time for Recurrent Neural Networks, Arxiv
  39. A. van der Oord et al., Pixel Recurrent Neural Networks, 2016, Arxiv
  40. C. Doersch, A Tutorial on Variational Autoencoders, 2016, Arxiv
  41. Ian Goodfellow, NIPS 2016 Tutorial: Generative Adversarial Networks, 2016, Arxiv
  42. Arjovsky et al, Wasserstein GAN, 2017, Arxiv
  43. T. White, Sampling Generative Network, NIPS 2016, Arxiv

Last modified: Wednesday, 20 February 2019, 11:56 PM