Culturing evolution strategies to support the exploration of novel environments by an intelligent robotic agent

1998 ◽  
pp. 219-228
Author(s):  
Chan-Jin Chung ◽  
Robert G. Reynolds
Keyword(s):  
Evolution Strategies ◽  
Novel Environments ◽  
Robotic Agent

Selecting landmarks in novel environments

2009 ◽  
Author(s):  
Jared E. Miller ◽  
Laura A. Carlson
Keyword(s):  
Novel Environments

scholarly journals Long-Term Experimental Evolution in Escherichia coli. IV. Targets of Selection and the Specificity of Adaptation

Genetics ◽  
1996 ◽  
Vol 143 (1) ◽  
pp. 15-26 ◽  
Author(s):  
Michael Travisano ◽  
Richard E Lenski
Keyword(s):  
Escherichia Coli ◽  
Experimental Evolution ◽  
Common Ancestor ◽  
The Novel ◽  
Physiological Mechanisms ◽  
Outer Membranes ◽  
Limiting Nutrient ◽  
Novel Environments ◽  
Uptake Mechanisms

Abstract This study investigates the physiological manifestation of adaptive evolutionary change in 12 replicate populations of Escherichia coli that were propagated for 2000 generations in a glucose-limited environment. Representative genotypes from each population were assayed for fitness relative to their common ancestor in the experimental glucose environment and in 11 novel single-nutrient environments. After 2000 generations, the 12 derived genotypes had diverged into at least six distinct phenotypic classes. The nutrients were classified into four groups based upon their uptake physiology. All 12 derived genotypes improved in fitness by similar amounts in the glucose environment, and this pattern of parallel fitness gains was also seen in those novel environments where the limiting nutrient shared uptake mechanisms with glucose. Fitness showed little or no consistent improvement, but much greater genetic variation, in novel environments where the limiting nutrient differed from glucose in its uptake mechanisms. This pattern of fitness variation in the novel nutrient environments suggests that the independently derived genotypes adapted to the glucose environment by similar, but not identical, changes in the physiological mechanisms for moving glucose across both the inner and outer membranes.

scholarly journals Evolution-Strategies-Driven Optimization on Secure and Reconfigurable Interconnection PUF Networks

Electronics ◽  
2021 ◽  
Vol 10 (5) ◽  
pp. 537
Author(s):  
Hongxiang Gu ◽  
Miodrag Potkonjak
Keyword(s):  
Machine Learning ◽  
State Of The Art ◽  
Evolution Strategies ◽  
Light Weight ◽  
Network Configuration ◽  
System Complexity ◽  
Physical Unclonable Functions ◽  
Output Stability ◽  
Output Mapping ◽  
Weight Design

Physical Unclonable Functions (PUFs) are known for their unclonability and light-weight design. However, several known issues with state-of-the-art PUF designs exist including vulnerability against machine learning attacks, low output randomness, and low reliability. To address these problems, we present a reconfigurable interconnected PUF network (IPN) design that significantly strengthens the security and unclonability of strong PUFs. While the IPN structure itself significantly increases the system complexity and nonlinearity, the reconfiguration mechanism remaps the input–output mapping before an attacker could collect sufficient challenge-response pairs (CRPs). We also propose using an evolution strategies (ES) algorithm to efficiently search for a network configuration that is capable of producing random and stable responses. The experimental results show that applying state-of-the-art machine learning attacks result in less than 53.19% accuracy for single-bit output prediction on a reconfigurable IPN with random configurations. We also show that, when applying configurations explored by our proposed ES method instead of random configurations, the output randomness is significantly improved by 220.8% and output stability by at least 22.62% in different variations of IPN.

Sign in / Sign up

Export Citation Format

Share Document