1 Nonlinear models for Natural Image Statistics Urs Köster & Aapo Hyvärinen University of Helsinki.

Slides:



Advertisements
Similar presentations
The Helmholtz Machine P Dayan, GE Hinton, RM Neal, RS Zemel
Advertisements

V1 Physiology. Questions Hierarchies of RFs and visual areas Is prediction equal to understanding? Is predicting the mean responses enough? General versus.
Chapter 2.
Neural Network Models in Vision Peter Andras
E ffi cient Coding: From Retina Ganglion Cells To V2 Cells Honghao Shan Garrison W. Cottrell The Temporal Dynamics of Learning Center Gary's Unbelievable.
HMAX Models Architecture Jim Mutch March 31, 2010.
Tiled Convolutional Neural Networks TICA Speedup Results on the CIFAR-10 dataset Motivation Pretraining with Topographic ICA References [1] Y. LeCun, L.
Patch to the Future: Unsupervised Visual Prediction
黃文中 Preview 2 3 The Saliency Map is a topographically arranged map that represents visual saliency of a corresponding visual scene. 4.
A saliency map model explains the effects of random variations along irrelevant dimensions in texture segmentation and visual search Li Zhaoping, University.
Learning Convolutional Feature Hierarchies for Visual Recognition
Bayesian Robust Principal Component Analysis Presenter: Raghu Ranganathan ECE / CMR Tennessee Technological University January 21, 2011 Reading Group (Xinghao.
Unsupervised Learning With Neural Nets Deep Learning and Neural Nets Spring 2015.
Application of Statistical Techniques to Neural Data Analysis Aniket Kaloti 03/07/2006.
Information Theory and Learning
ICA Alphan Altinok. Outline  PCA  ICA  Foundation  Ambiguities  Algorithms  Examples  Papers.
Linear Algebra and Image Processing
Sparse Coding Arthur Pece Outline Generative-model-based vision Linear, non-Gaussian, over-complete generative models The penalty method.
Another viewpoint: V1 cells are spatial frequency filters
Cortical Receptive Fields Using Deep Autoencoders Work done as a part of CS397 Ankit Awasthi (Y8084) Supervisor: Prof. H. Karnick.
CSC2535: Computation in Neural Networks Lecture 11: Conditional Random Fields Geoffrey Hinton.
1 / 41 Inference and Computation with Population Codes 13 November 2012 Inference and Computation with Population Codes Alexandre Pouget, Peter Dayan,
The search for organizing principles of brain function Needed at multiple levels: synapse => cell => brain area (cortical maps) => hierarchy of areas.
INDEPENDENT COMPONENT ANALYSIS OF TEXTURES based on the article R.Manduchi, J. Portilla, ICA of Textures, The Proc. of the 7 th IEEE Int. Conf. On Comp.
Independence of luminance and contrast in natural scenes and in the early visual system Valerio Mante, Robert A Frazor, Vincent Bonin, Wilson S Geisler,
Low Level Visual Processing. Information Maximization in the Retina Hypothesis: ganglion cells try to transmit as much information as possible about the.
2 2  Background  Vision in Human Brain  Efficient Coding Theory  Motivation  Natural Pictures  Methodology  Statistical Characteristics  Models.
Neuroinformatics Aapo Hyvärinen Professor, Group Leader.
Lecture 2b Readings: Kandell Schwartz et al Ch 27 Wolfe et al Chs 3 and 4.
FMRI Methods Lecture7 – Review: analyses & statistics.
Projects: 1.Predictive coding in balanced spiking networks (Erwan Ledoux). 2.Using Canonical Correlation Analysis (CCA) to analyse neural data (David Schulz).
On Natural Scenes Analysis, Sparsity and Coding Efficiency Redwood Center for Theoretical Neuroscience University of California, Berkeley Mind, Brain.
Fields of Experts: A Framework for Learning Image Priors (Mon) Young Ki Baik, Computer Vision Lab.
Csc Lecture 8 Modeling image covariance structure Geoffrey Hinton.
CSC2515: Lecture 7 (post) Independent Components Analysis, and Autoencoders Geoffrey Hinton.
Lecture 2: Statistical learning primer for biologists
CSC321 Lecture 5 Applying backpropagation to shape recognition Geoffrey Hinton.
Convolutional Restricted Boltzmann Machines for Feature Learning Mohammad Norouzi Advisor: Dr. Greg Mori Simon Fraser University 27 Nov
Markov Random Fields in Vision
Texture Analysis and Synthesis. Texture Texture: pattern that “looks the same” at all locationsTexture: pattern that “looks the same” at all locations.
Jianchao Yang, John Wright, Thomas Huang, Yi Ma CVPR 2008 Image Super-Resolution as Sparse Representation of Raw Image Patches.
Statistical Modeling and Learning in Vision --- cortex-like generative models Ying Nian Wu UCLA Department of Statistics JSM, August 2010.
Deep Learning Overview Sources: workshop-tutorial-final.pdf
Independent Component Analysis features of Color & Stereo images Authors: Patrik O. Hoyer Aapo Hyvarinen CIS 526: Neural Computation Presented by: Ajay.
CSC2535: Computation in Neural Networks Lecture 7: Independent Components Analysis Geoffrey Hinton.
CSC2535: Lecture 4: Autoencoders, Free energy, and Minimum Description Length Geoffrey Hinton.
Bayesian inference & visual processing in the brain
Biointelligence Laboratory, Seoul National University
Regression Analysis AGEC 784.
Deep Learning Amin Sobhani.
Energy models and Deep Belief Networks
CSC321: Neural Networks Lecture 22 Learning features one layer at a time Geoffrey Hinton.
Spontaneous activity in V1: a probabilistic framework
Article Review Todd Hricik.
The general linear model and Statistical Parametric Mapping
Nonparametric Semantic Segmentation
Towards Understanding the Invertibility of Convolutional Neural Networks Anna C. Gilbert1, Yi Zhang1, Kibok Lee1, Yuting Zhang1, Honglak Lee1,2 1University.
Nonlinear processing in LGN neurons
Perception: Structures
Texture.
The general linear model and Statistical Parametric Mapping
Outline Texture modeling - continued Julesz ensemble.
Binocular Disparity and the Perception of Depth
M/EEG Statistical Analysis & Source Localization
Discovery of Hidden Structure in High-Dimensional Data
Neural Network Models in Vision
Reuben Feinman Research advised by Brenden Lake
NON-NEGATIVE COMPONENT PARTS OF SOUND FOR CLASSIFICATION Yong-Choon Cho, Seungjin Choi, Sung-Yang Bang Wen-Yi Chu Department of Computer Science &
How to win big by thinking straight about relatively trivial problems
Goodfellow: Chapter 14 Autoencoders
Presentation transcript:

1 Nonlinear models for Natural Image Statistics Urs Köster & Aapo Hyvärinen University of Helsinki

2 Overview  A two-layer model learns complex cell pooling  A horizontal product model for Contrast Gain Control  A Markov Random Field generalizes ICA to large images (“convolutional ICA”)  Things I would really like to know 2

Limitations of linear image models Natural images have complex structure Linear models ignore much of the rich interactions between units Interactions related to modelling the dependencies of linear components Variance dependencies are a particularly obvious nonlinear dependency Here, we consider Models with (complex cell) pooling of linear filter outputs - hierarchical models Model dependencies by gain control on the pixel level Schwartz & Simoncelli 2001

4 A two-layer model estimated with score matching learns complex cell -like receptive fields

5 Two-layer model estimated with score matching Define an energy based model of the form Squaring the outputs of linear filters Second layer linear transform v are free, as opposed to topographic ICA (Hyvärinen and Hoyer 2001) Non-normalized model: needs either score matching or MCMC methods Related to Osindero & Hinton (2006), Karklin and Lewicki (2005)

6 Two-layer model: Results Some pooling patterns The second layer learns to pool over units with similar location and orientation, but different spatial phase Learns the energy model of Complex Cells without any assumptions on the pooling (but constrained non-negative) Estimating W and V simultaneously leads to a better optimum and more phase invariance of the higher order units

7 Learning to perform local gain control with a horizontal product model

Gain Control in Physiology Gain control is common throughout the brain The divisive normalization model for primary visual cortex Previous work has analyzed the effect of gain control on the cortical level We use a statistical model for gain control on the LGN level Important effects on subsequent cortical processing Normalization model Carandini and Heeger,1994

9 Multiplicative interactions Data is described by element-wise multiplying outputs of sub-models Combine aspects of a stimulus generated by separate mechanisms Horizontal layers Two parallel streams or layers on one level of the hierarchy Unrelated aspects of the stimulus are generated separately Observed data is generated by combining all the sub-models A horizontal network model:

10 The model Definition of the model: Likelihood: Constraints: B and t are non-negative, low dimension, W invertible g(.) is log cosh (logistic distribution) t has an L1 sparseness prior 10

Horizontal product model: results First layer W 4 units in B First layer W 16 units in B

Emerging Contrast Gain Control Reconstruction from As only True image patches Modulation from Bt Emergence of a contrast map in the second layer It performs Contrast Gain Control on the LGN level (rather than on filter outputs) Effectively performing some kind of divisive normalization locally The model can be written as

13 The “big” picture: A Markov Random Field generalizes ICA to images of arbitrary size

14 Markov Random Field Goal: Define probabilities for whole images rather than small patches A MRF uses a convolution to analyze large images with small filters Estimating the optimal filters difficult: the model cannot be normalized Estimation using score matching

15 MRF: Model estimation The energy (neg. log pdf) is We can rewrite the convolution where x i are all possible patches from the image, w k are the different filters Non-normalized model: We can use score matching The MRF is roughly equivalent to overcomplete ICA with filters copied in all possible locations.

16 MRF: Results We can estimate MRF filters of size 12x12 pixels (much larger than previous work, e.g. 5x5) This is possible from 23x23 pixel ‘images’, but the filters generalize to images of arbitrary size This is possible because all possible overlaps are accounted for Filters similar to ICA, but less localized (since they need to explain more of the surrounding patch?) Possible applications in denoising and filling-in 16

17 Things I would like to know

18 Things I would like to know (1) What is the role of unsupervised learning from passively observed data (“Pure cognition”)? What about unsupervised learning in an agent (e.g. Mahadevan et al 2007) Action and perception coupled Cf. reinforcement learning

19 Things I would like to know (2) If we admit pure cognition, what is the goal of statistical modelling in the brain? What is the proper theoretical framework? Coding (information theory) or Bayesian inference Hardly sparsity (statistically) or independence, I think Or is it just about reducing metabolic costs? Sparse coding Minimum wiring length

20 More things I would like to know What is V2 doing? What are meaningful nonlinearities to use? Squaring in the first layer, perhaps.. Can we just use the same logistic everywhere? [This space for sale]