Balancing the Costs and Benefits of Learning Ability

We study the costs and benefits of plasticity by evolving agents in environments with different rates of environmental change. Evolution allows both hard-coded strategies and learned strategies, with learning rates varying throughout life. We observe a range of change rates where the balance of costs and benefits are just right for evolving learning. Inside this range, we see two separate strategies evolve: lifelong plasticity and sensitive periods of plasticity. Sensitive periods of plasticity are found to reduce the learning cost while retaining the benefits of learning. This affects the evolutionary process, by limiting genetic assimilation of learned characteristics, making agents able to remain adaptive after relatively long periods of environmental stability.

[1]  R. Anderson,et al.  Learning and evolution: a quantitative genetics approach. , 1995, Journal of theoretical biology.

[2]  Geoffrey E. Hinton,et al.  How Learning Can Guide Evolution , 1996, Complex Syst..

[3]  D. Hubel,et al.  The period of susceptibility to the physiological effects of unilateral eye closure in kittens , 1970, The Journal of physiology.

[4]  J. Bullinaria.  From biological models to the evolution of robot control systems , 2003, Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[5]  Michael L. Littman,et al.  Simulations combining evolution and learning , 1996 .

[6]  Giles Mayley.  The Evolutionary Cost of Learning , 1996 .

[7]  Maja J. Matarić,et al.  The Evolutionary Cost of Learning , 1996 .

[8]  Marcus W Feldman,et al.  Carving the cognitive niche: optimal learning strategies in homogeneous and heterogeneous environments. , 2003, Journal of theoretical biology.

[9]  A. D. Bradshaw,et al.  Evolutionary Significance of Phenotypic Plasticity in Plants , 1965 .

[10]  Peter M. Todd,et al.  Exploring adaptive agency II: simulating the evolution of associative learning , 1991 .

[11]  Jeffrey L. Elman,et al.  Learning and Evolution in Neural Networks , 1994, Adapt. Behav..

[12]  Petr E. Komers,et al.  Behavioural plasticity in variable environments , 1997 .

[13]  J. Baldwin.  A New Factor in Evolution , 1896, The American Naturalist.

[14]  Giles Mayley,et al.  Landscapes, Learning Costs, and Genetic Assimilation , 1996, Evolutionary Computation.

[15]  Dario Floreano,et al.  Evolution of Plastic Control Networks , 2001, Auton. Robots.

[16]  Janet Wiles,et al.  Stability and task complexity: a neural network model of genetic assimilation , 2002 .

[17]  Peter D. Turney.  Myths and Legends of the Baldwin Effect , 2002, ICML 2002.

[18]  E. Knudsen.  Sensitive Periods in the Development of the Brain and Behavior , 2004, Journal of Cognitive Neuroscience.

[19]  D. Stephens,et al.  Components of change in the evolution of learning and unlearned preference , 2009, Proceedings of the Royal Society B: Biological Sciences.

[20]  Takahiro Sasaki,et al.  Evolving Learnable Neural Networks Under Changing Environments with Various Rates of Inheritance of Acquired Characters: Comparison of Darwinian and Lamarckian Evolution , 1999, Artificial Life.

[21]  S. Kirby,et al.  The evolution of incremental learning: language, development and critical periods , 1997 .