Evolution of visually guided behavior in artificial agents

Recent work on brightness, color, and form has suggested that human visual percepts represent the probable sources of retinal images rather than stimulus features as such. Here we investigate the plausibility of this empirical concept of vision by allowing autonomous agents to evolve in virtual environments based solely on the relative success of their behavior. The responses of evolved agents to visual stimuli indicate that fitness improves as the neural network control systems gradually incorporate the statistical relationship between projected images and behavior appropriate to the sources of the inherently ambiguous images. These results: (1) demonstrate the merits of a wholly empirical strategy of animal vision as a means of contending with the inverse optics problem; (2) argue that the information incorporated into biological visual processing circuitry is the relationship between images and their probable sources; and (3) suggest why human percepts do not map neatly onto physical reality.

[1]  Jeffrey L. Krichmar,et al.  Evolutionary robotics: The biology, intelligence, and technology of self-organizing machines , 2001, Complex..

[2]  D. Floreano,et al.  Evolutionary Robotics: The Biology,Intelligence,and Technology , 2000 .

[3]  D. Hubel,et al.  Receptive fields and functional architecture of monkey striate cortex , 1968, The Journal of physiology.

[4]  David J. Field,et al.  Emergence of simple-cell receptive field properties by learning a sparse code for natural images , 1996, Nature.

[5]  P. König,et al.  A Model of the Ventral Visual System Based on Temporal Stability and Local Memory , 2006, PLoS biology.

[6]  D. Purves,et al.  An empirical explanation of color contrast. , 2000, Proceedings of the National Academy of Sciences of the United States of America.

[7]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[8]  J. Gibson.  The Ecological Approach to Visual Perception , 1979 .

[9]  Thomas Bäck,et al.  Evolutionary algorithms in theory and practice - evolution strategies, evolutionary programming, genetic algorithms , 1996 .

[10]  Francesco Mondada,et al.  Evolution of homing navigation in a real mobile robot , 1996, IEEE Trans. Syst. Man Cybern. Part B.

[11]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

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

[13]  Dale Purves,et al.  The statistical structure of natural light patterns determines perceived light intensity. , 2004, Proceedings of the National Academy of Sciences of the United States of America.

[14]  Jean Ponce,et al.  Computer Vision: A Modern Approach , 2002 .

[15]  S. Nolfi,et al.  Evolving Robots Able to Visually Discriminate Between Objects with Different Size , 2003 .

[16]  Terrence J. Sejnowski,et al.  The “independent components” of natural scenes are edge filters , 1997, Vision Research.

[17]  Christopher M. Bishop,et al.  Neural networks for pattern recognition , 1995 .

[18]  David J. Fleet,et al.  Probabilistic Models of the Brain : Perception and Neural Function , 2001 .

[19]  D. Purves,et al.  Why we see what we do : an empirical theory of vision , 2003 .

[20]  Dale Purves,et al.  Natural scene statistics as the universal basis of color context effects , 2003, Proceedings of the National Academy of Sciences of the United States of America.

[21]  T. Sejnowski,et al.  Color opponency is an efficient representation of spectral properties in natural scenes , 2002, Vision Research.

[22]  Inman Harvey,et al.  Artificial evolution of visual control systems for robots , 1997 .

[23]  Berthold K. P. Horn.  Obtaining shape from shading information , 1989 .

[24]  Dale Purves,et al.  Range image statistics can explain the anomalous perception of length , 2002, Proceedings of the National Academy of Sciences of the United States of America.

[25]  Stefano Nolfi,et al.  Evolving Visually-Guided Robots Able to Discriminate Between Different Landmarks , 2000 .

[26]  M. Wertheimer.  A source book of Gestalt psychology. , 1939 .

[27]  Francesco Mondada,et al.  Evolutionary neurocontrollers for autonomous mobile robots , 1998, Neural Networks.

[28]  H. Helmholtz.  Helmholtz's Treatise on Physiological Optics , 1963 .

[29]  Tom M. C. Smith,et al.  Blurred Vision: Simulation-Reality Transfer of a Visually Guided Robot , 1998, EvoRobot.

[30]  W. PEDDIE,et al.  Helmholtz's Treatise on Physiological Optics , 1926, Nature.

[31]  Tai Sing Lee,et al.  Statistical correlations between two-dimensional images and three-dimensional structures in natural scenes. , 2003, Journal of the Optical Society of America. A, Optics, image science, and vision.

[32]  Ralf Salomon,et al.  Increasing Adaptivity through Evolution Strategies , 1996 .

[33]  London,et al.  The Ames Demonstrations in Perception , 1953 .

[34]  D. Purves,et al.  Size Contrast and Assimilation Explained by the Statistics of Natural Scene Geometry , 2004, Journal of Cognitive Neuroscience.

[35]  Dale Purves,et al.  Perceiving the intensity of light. , 2004, Psychological review.

[36]  Pattie Maes,et al.  Increasing Adaptivity through Evolution Strategies , 1996 .

[37]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[38]  D. Fitzpatrick,et al.  Orientation Selectivity and the Arrangement of Horizontal Connections in Tree Shrew Striate Cortex , 1997, The Journal of Neuroscience.

[39]  W. Richards,et al.  Perception as Bayesian Inference , 2008 .

[40]  Dale Purves,et al.  A statistical explanation of visual space , 2003, Nature Neuroscience.

[41]  Dale Purves,et al.  Natural-scene geometry predicts the perception of angles and line orientation. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[42]  Fuhui Long,et al.  Spectral statistics in natural scenes predict hue, saturation, and brightness. , 2006, Proceedings of the National Academy of Sciences of the United States of America.

[43]  D. Hubel,et al.  Receptive fields, binocular interaction and functional architecture in the cat's visual cortex , 1962, The Journal of physiology.

[44]  Rolf Pfeifer,et al.  Understanding intelligence , 2020, Inequality by Design.

[45]  Robert A. Dain.  Developing Mobile Robot Wall-Following Algorithms Using Genetic Programming , 2004, Applied Intelligence.

[46]  Dale Purves,et al.  Comparison of Bayesian and empirical ranking approaches to visual perception. , 2006, Journal of theoretical biology.

[47]  Jeffrey S. Perry,et al.  Edge co-occurrence in natural images predicts contour grouping performance , 2001, Vision Research.