Incremental Slow Feature Analysis: Adaptive and Episodic Learning from High-Dimensional Input Streams

Slow Feature Analysis (SFA) extracts features representing the underlying causes of changes within a temporally coherent high-dimensional raw sensory input signal. Our novel incremental version of SFA (IncSFA) combines incremental Principal Components Analysis and Minor Components Analysis. Unlike standard batch-based SFA, IncSFA adapts along with non-stationary environments, is amenable to episodic training, is not corrupted by outliers, and is covariance-free. These properties make IncSFA a generally useful unsupervised preprocessor for autonomous learning agents and robots. In IncSFA, the CCIPCA and MCA updates take the form of Hebbian and anti-Hebbian updating, extending the biological plausibility of SFA. In both single node and deep network versions, IncSFA learns to encode its input streams (such as high-dimensional video) by informative slow features representing meaningful abstract environmental properties. It can handle cases where batch SFA fails.

[1]  Peter Földiák,et al.  Learning Invariance from Transformation Sequences , 1991, Neural Comput..

[2]  Niko Wilbert,et al.  Slow feature analysis , 2011, Scholarpedia.

[3]  Jürgen Schmidhuber,et al.  Formal Theory of Creativity, Fun, and Intrinsic Motivation (1990–2010) , 2010, IEEE Transactions on Autonomous Mental Development.

[4]  Geoffrey E. Hinton.  Connectionist Learning Procedures , 1989, Artif. Intell..

[5]  I. Miller.  Probability, Random Variables, and Stochastic Processes , 1966 .

[6]  Graeme Mitchison,et al.  Removing Time Variation with the Anti-Hebbian Differential Synapse , 1991, Neural Computation.

[7]  J. Weng,et al.  Convergence Analysis of Complementary Candid Incremental Principal Component Analysis ∗ , 2001 .

[8]  H. Abut,et al.  Vector Quantization , 1990 .

[9]  Qin Lin,et al.  A unified algorithm for principal and minor components extraction , 1998, Neural Networks.

[10]  Peter Dayan,et al.  Theoretical Neuroscience: Computational and Mathematical Modeling of Neural Systems , 2001 .

[11]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[12]  Jürgen Schmidhuber,et al.  Learning Complex, Extended Sequences Using the Principle of History Compression , 1992, Neural Computation.

[13]  Zhang Yi,et al.  Convergence analysis of a simple minor component analysis algorithm , 2007, Neural Networks.

[14]  E. Rolls.  Spatial view cells and the representation of place in the primate hippocampus , 1999, Hippocampus.

[15]  Robert A. Legenstein,et al.  Reinforcement Learning on Slow Features of High-Dimensional Input Streams , 2010, PLoS Comput. Biol..

[16]  Honglak Lee,et al.  Unsupervised feature learning for audio classification using convolutional deep belief networks , 2009, NIPS.

[17]  Jürgen Schmidhuber,et al.  Unsupervised Learning in LSTM Recurrent Neural Networks , 2001, ICANN.

[18]  Terrence J. Sejnowski,et al.  Slow Feature Analysis: Unsupervised Learning of Invariances , 2002, Neural Computation.

[19]  I. Jolliffe.  Principal Component Analysis , 2002 .

[20]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..

[21]  Jürgen Schmidhuber.  Neural Predictors for Detecting and Removing Redundant Information , 2000 .

[22]  Jürgen Schmidhuber,et al.  Learning Factorial Codes by Predictability Minimization , 1992, Neural Computation.

[23]  Stefanie N. Lindstaedt,et al.  Comparison of two Unsupervised Neural Network Models for Redundancy Reduction , 1993 .

[24]  R U Muller,et al.  Head-direction cells recorded from the postsubiculum in freely moving rats. I. Description and quantitative analysis , 1990, The Journal of neuroscience : the official journal of the Society for Neuroscience.

[25]  T. Hafting,et al.  Microstructure of a spatial map in the entorhinal cortex , 2005, Nature.

[26]  Dezhong Peng,et al.  A New Algorithm for Sequential Minor Component Analysis , 2006 .

[27]  Erkki Oja,et al.  Modified Hebbian learning for curve and surface fitting , 1992, Neural Networks.

[28]  R. Muller,et al.  Head-direction cells recorded from the postsubiculum in freely moving rats. II. Effects of environmental manipulations , 1990, The Journal of neuroscience : the official journal of the Society for Neuroscience.

[29]  Pierre Comon.  Independent component analysis - a new concept? signal processing , 1994 .

[30]  Juyang Weng,et al.  Candid Covariance-Free Incremental Principal Component Analysis , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[31]  Juha Karhunen,et al.  A Unified Neural Bigradient Algorithm for robust PCA and MCA , 1996, Int. J. Neural Syst..

[32]  Niko Wilbert,et al.  Modular Toolkit for Data Processing (MDP): A Python Data Processing Framework , 2008, Frontiers Neuroinformatics.

[33]  Carl Doersch,et al.  Temporal Continuity Learning for Convolutional Deep Belief Networks , 2010 .

[34]  Terence D. Sanger,et al.  Optimal unsupervised learning in a single-layer linear feedforward neural network , 1989, Neural Networks.

[35]  E. Oja,et al.  On stochastic approximation of the eigenvectors and eigenvalues of the expectation of a random matrix , 1985 .

[36]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[37]  J. Meigs,et al.  WHO Technical Report , 1954, The Yale Journal of Biology and Medicine.

[38]  Athanasios Papoulis,et al.  Probability, Random Variables and Stochastic Processes , 1965 .

[39]  Yoshua Bengio,et al.  Slow, Decorrelated Features for Pretraining Complex Cell-like Networks , 2009, NIPS.

[40]  Vwani P. Roychowdhury,et al.  Algorithms for accelerated convergence of adaptive PCA , 2000, IEEE Trans. Neural Networks Learn. Syst..

[41]  Jürgen Schmidhuber,et al.  Incremental Slow Feature Analysis , 2011, IJCAI.

[42]  Maja J. Mataric,et al.  A spatio-temporal extension to Isomap nonlinear dimension reduction , 2004, ICML.

[43]  H Barlow,et al.  Redundancy reduction revisited , 2001, Network.

[44]  Laurenz Wiskott,et al.  Slowness and Sparseness Lead to Place, Head-Direction, and Spatial-View Cells , 2007, PLoS Comput. Biol..

[45]  F. W. Kellaway,et al.  Advanced Engineering Mathematics , 1969, The Mathematical Gazette.

[46]  Jürgen Schmidhuber,et al.  Learning Unambiguous Reduced Sequence Descriptions , 1991, NIPS.

[47]  S. Grossberg.  How does a brain build a cognitive code , 1980 .

[48]  Angelo Cangelosi,et al.  An open-source simulator for cognitive robotics research: the prototype of the iCub humanoid robot simulator , 2008, PerMIS.

[49]  Jürgen Schmidhuber,et al.  AutoIncSFA and vision-based developmental learning for humanoid robots , 2011, 2011 11th IEEE-RAS International Conference on Humanoid Robots.

[50]  E. Oja.  Simplified neuron model as a principal component analyzer , 1982, Journal of mathematical biology.

[51]  J. O'Keefe,et al.  The hippocampus as a spatial map. Preliminary evidence from unit activity in the freely-moving rat. , 1971, Brain research.

[52]  Geoffrey E. Hinton.  Training Products of Experts by Minimizing Contrastive Divergence , 2002, Neural Computation.

[53]  Laurenz Wiskott.  Estimating Driving Forces of Nonstationary Time Series with Slow Feature Analysis Laurenz Wiskott Institute for Theoretical Biology , 2003 .

[54]  Jürgen Schmidhuber,et al.  Sequential Constant Size Compressors for Reinforcement Learning , 2011, AGI.

[55]  Erkki Oja,et al.  Principal components, minor components, and linear neural networks , 1992, Neural Networks.

[56]  S.-I. Amari,et al.  Neural theory of association and concept-formation , 1977, Biological Cybernetics.

[57]  Shun-ichi Amari,et al.  Sequential Extraction of Minor Components , 2001, Neural Processing Letters.

[58]  Juyang Weng,et al.  Optimal In-Place Learning and the Lobe Component Analysis , 2006, The 2006 IEEE International Joint Conference on Neural Network Proceedings.