Skip to Main Content

Table 8:
Datasets and performance metrics of x-NEAT methods belonging to cluster 1.
DatasetsPerformance metrics

Deceptive Landscape

 

(NoveltyNEAT, FNS-NEATFields, NS-FE-CPPN-NEAT, MAP-Elites CPPN, NEAT-LSTM-IM)

 

Maze navigation (Lehman and Stanley, 2011a; Inden et al., 2013)

Locomotion of a 3D biped robot (Lehman and Stanley, 2011a)

The pole balancing (Inden et al., 2013)

Distinction of 4 visual patterns (Inden et al., 2013)

Distinction of 2 visual patterns with variable position (Inden et al., 2013)

Robot locomotion (Tarapore et al., 2016)

Evolving the morphology and the locomotion of soft robots (Methenitis et al., 2015)

Sequence classification and recall in T-maze (Rawal and Miikkulainen, 2016)

 

Ratio of successful runs (Rawal and Miikkulainen, 2016)

Average number of evaluations (Lehman and Stanley, 2011a)

Size of networks (Lehman and Stanley, 2011a)

Achieved fitness divided by the number of generations (Inden et al., 2013)

Global Fitness (Tarapore et al., 2016)

Coverage (Tarapore et al., 2016)

Global reliability (Tarapore et al., 2016)

Precision (Tarapore et al., 2016)

Fitness value (Methenitis et al., 2015)

 

Uncertain Landscape (DynNEAT)

 

Function approximation problems (Krčah, 2012)

 

Fitness value (Krčah, 2012)

Ratio of successful runs (Krčah, 2012)

Number of evolved species (Krčah, 2012)

 

Open-Ended Problems

 

(Coevolutionary NEAT, LAPCA-NEAT, nnrg.hazel, NoveltyNEAT)

 

Robots' duel (Stanley and Miikkulainen, 2004)

Pong game (Monroy et al., 2006; Winter, 2005)

General Game Playing (Reisinger et al., 2007; Genesereth et al., 2005):

i. connect four

ii. chinese checkers

iii. crisscross

iv. blocker

v. two-board tic-tac-toe

2D maze navigation task (Lehman and Stanley, 2011a)

Locomotion of a 3D biped robot (Lehman and Stanley, 2011a)

 

Dominance metrics (Stanley and Miikkulainen, 2004; Monroy et al., 2006; Reisinger et al., 2007)

Generation of achieved dominance (Stanley and Miikkulainen, 2004)

Size of networks (Stanley and Miikkulainen, 2004; Lehman and Stanley, 2011a)

Average number of wins/ties/losses (Monroy et al., 2006)

Game's score (Reisinger et al., 2007)

Memory's size (Monroy et al., 2006)

Number of generations for which a set of dominated strategies is undominated (Reisinger et al., 2007)

Average number of evaluations (Lehman and Stanley, 2011a)

 

Multiple Objectives

 

(MO-NEAT, MM-NEAT)

 

Communication's optimization between a sensor node and a base station (Haggett and Chu, 2009)

Anomaly detection in time series data (Haggett and Chu, 2009)

Ms Pac Man (Schrum and Miikkulainen, 2016)

 

Fitness Value (Schrum and Miikkulainen, 2016)

Percentage of True Positives and False Positives (Haggett and Chu, 2009)

Post-learning score (Schrum and Miikkulainen, 2016)

Percentage of time of mostly used module (Schrum and Miikkulainen, 2016)

Data specific metrics (Haggett and Chu, 2009):

-Percentage of transmitted packets

 

Irrelevant or Redundant Features

 

(FS-NEAT, FD-NEAT, IFSE-NEAT, Layered NEAT, Phased NEAT, PFS-NEAT)

 

Robot Auto Racing Simulator (RARS) (Whiteson et al., 2005; Tan et al., 2009; Loscalzo et al., 2015; Timin, 1995; Wright et al., 2012)

(2/5) XOR (Tan et al., 2009; Wang et al., 2013)

The concentric spirals (Tan et al., 2009; Potter and Jong, 2000)

The surface plots (Tan et al., 2009)

Marcellus Shale lithofacies prediction (Wang et al., 2013)

Lung Nodule classification (Tan et al., 2013)

Double pole balancing (Loscalzo et al., 2015)

 

Fitness value (Whiteson et al., 2005; Tan et al., 2009; Wright et al., 2012; Wang et al., 2013; Tan et al., 2013; Loscalzo et al., 2015)

Number of generations (Whiteson et al., 2005; Tan et al., 2009)

Computational time (Wang et al., 2013)

Size of networks (Whiteson et al., 2005; Tan et al., 2009; Wang et al., 2013; Tan et al., 2013)

Feature selection metrics:

-Frequency of selecting relevant inputs (Tan et al., 2009)

-Sum of absolute weights of input connections (Tan et al., 2009)

-Ratio of relevant--irrelevant features (Wright et al., 2012; Loscalzo et al., 2015)

-Number of selected features (Loscalzo et al., 2015)

 

Online Evolution (Online NEAT+Q, KO-NEAT, Online NEAT, Online rtNEAT, odNEAT, odNEATv2)

 

Real Time Evolution (rtNEAT, rtNEATv2, cgNEAT, Online rtNEAT)

 

The mountain car task (Whiteson and Stone, 2006a; Boyan and Moore, 1995)

Server Job Scheduling (Whiteson and Stone, 2006a)

Keepaway Soccer (Zhao et al., 2007; Stone et al., 2005)

NERO video game (Stanley et al., 2005; D'Silva et al., 2005; Stanley et al., 2006)

Galactic Arms Race (Hastings et al., 2009; Games, 2010)

The Open Racing Car Simulator (TORCS) (Whiteson and Stone, 2006b; Reeder et al., 2008; Cardamone et al., 2010; Stanley et al., 2016)

Multiagent aggregation task (Silva et al., 2015)

Multiagent integrated navigation & obstacle avoidance (Silva et al., 2015)

Foraging tasks (Silva et al., 2016)

Multiagent Phototaxis task (Silva et al., 2015, 2016)

 

Uniform Moving Average Score (Whiteson and Stone, 2006a)

Utility function (Whiteson and Stone, 2006a; Walsh et al., 2004)

Fitness Value (Zhao et al., 2007; Silva et al., 2015, 2016)

Number of evaluations (Zhao et al., 2007; Silva et al., 2015)

Size of networks (Silva et al., 2015)

Data specific metrics:

-Subjective metric of user's perception (Stanley et al., 2005)

-Users' statistics (Hastings et al., 2009)

-Remaining distance to the target (D'Silva et al., 2005)

-Portion of successful agents (D'Silva et al., 2005)

-Evolved content (Hastings et al., 2009)

-Average lap time in a car simulator (Cardamone et al., 2010)

Controllers' operation time (Silva et al., 2016)

 
DatasetsPerformance metrics

Deceptive Landscape

 

(NoveltyNEAT, FNS-NEATFields, NS-FE-CPPN-NEAT, MAP-Elites CPPN, NEAT-LSTM-IM)

 

Maze navigation (Lehman and Stanley, 2011a; Inden et al., 2013)

Locomotion of a 3D biped robot (Lehman and Stanley, 2011a)

The pole balancing (Inden et al., 2013)

Distinction of 4 visual patterns (Inden et al., 2013)

Distinction of 2 visual patterns with variable position (Inden et al., 2013)

Robot locomotion (Tarapore et al., 2016)

Evolving the morphology and the locomotion of soft robots (Methenitis et al., 2015)

Sequence classification and recall in T-maze (Rawal and Miikkulainen, 2016)

 

Ratio of successful runs (Rawal and Miikkulainen, 2016)

Average number of evaluations (Lehman and Stanley, 2011a)

Size of networks (Lehman and Stanley, 2011a)

Achieved fitness divided by the number of generations (Inden et al., 2013)

Global Fitness (Tarapore et al., 2016)

Coverage (Tarapore et al., 2016)

Global reliability (Tarapore et al., 2016)

Precision (Tarapore et al., 2016)

Fitness value (Methenitis et al., 2015)

 

Uncertain Landscape (DynNEAT)

 

Function approximation problems (Krčah, 2012)

 

Fitness value (Krčah, 2012)

Ratio of successful runs (Krčah, 2012)

Number of evolved species (Krčah, 2012)

 

Open-Ended Problems

 

(Coevolutionary NEAT, LAPCA-NEAT, nnrg.hazel, NoveltyNEAT)

 

Robots' duel (Stanley and Miikkulainen, 2004)

Pong game (Monroy et al., 2006; Winter, 2005)

General Game Playing (Reisinger et al., 2007; Genesereth et al., 2005):

i. connect four

ii. chinese checkers

iii. crisscross

iv. blocker

v. two-board tic-tac-toe

2D maze navigation task (Lehman and Stanley, 2011a)

Locomotion of a 3D biped robot (Lehman and Stanley, 2011a)

 

Dominance metrics (Stanley and Miikkulainen, 2004; Monroy et al., 2006; Reisinger et al., 2007)

Generation of achieved dominance (Stanley and Miikkulainen, 2004)

Size of networks (Stanley and Miikkulainen, 2004; Lehman and Stanley, 2011a)

Average number of wins/ties/losses (Monroy et al., 2006)

Game's score (Reisinger et al., 2007)

Memory's size (Monroy et al., 2006)

Number of generations for which a set of dominated strategies is undominated (Reisinger et al., 2007)

Average number of evaluations (Lehman and Stanley, 2011a)

 

Multiple Objectives

 

(MO-NEAT, MM-NEAT)

 

Communication's optimization between a sensor node and a base station (Haggett and Chu, 2009)

Anomaly detection in time series data (Haggett and Chu, 2009)

Ms Pac Man (Schrum and Miikkulainen, 2016)

 

Fitness Value (Schrum and Miikkulainen, 2016)

Percentage of True Positives and False Positives (Haggett and Chu, 2009)

Post-learning score (Schrum and Miikkulainen, 2016)

Percentage of time of mostly used module (Schrum and Miikkulainen, 2016)

Data specific metrics (Haggett and Chu, 2009):

-Percentage of transmitted packets

 

Irrelevant or Redundant Features

 

(FS-NEAT, FD-NEAT, IFSE-NEAT, Layered NEAT, Phased NEAT, PFS-NEAT)

 

Robot Auto Racing Simulator (RARS) (Whiteson et al., 2005; Tan et al., 2009; Loscalzo et al., 2015; Timin, 1995; Wright et al., 2012)

(2/5) XOR (Tan et al., 2009; Wang et al., 2013)

The concentric spirals (Tan et al., 2009; Potter and Jong, 2000)

The surface plots (Tan et al., 2009)

Marcellus Shale lithofacies prediction (Wang et al., 2013)

Lung Nodule classification (Tan et al., 2013)

Double pole balancing (Loscalzo et al., 2015)

 

Fitness value (Whiteson et al., 2005; Tan et al., 2009; Wright et al., 2012; Wang et al., 2013; Tan et al., 2013; Loscalzo et al., 2015)

Number of generations (Whiteson et al., 2005; Tan et al., 2009)

Computational time (Wang et al., 2013)

Size of networks (Whiteson et al., 2005; Tan et al., 2009; Wang et al., 2013; Tan et al., 2013)

Feature selection metrics:

-Frequency of selecting relevant inputs (Tan et al., 2009)

-Sum of absolute weights of input connections (Tan et al., 2009)

-Ratio of relevant--irrelevant features (Wright et al., 2012; Loscalzo et al., 2015)

-Number of selected features (Loscalzo et al., 2015)

 

Online Evolution (Online NEAT+Q, KO-NEAT, Online NEAT, Online rtNEAT, odNEAT, odNEATv2)

 

Real Time Evolution (rtNEAT, rtNEATv2, cgNEAT, Online rtNEAT)

 

The mountain car task (Whiteson and Stone, 2006a; Boyan and Moore, 1995)

Server Job Scheduling (Whiteson and Stone, 2006a)

Keepaway Soccer (Zhao et al., 2007; Stone et al., 2005)

NERO video game (Stanley et al., 2005; D'Silva et al., 2005; Stanley et al., 2006)

Galactic Arms Race (Hastings et al., 2009; Games, 2010)

The Open Racing Car Simulator (TORCS) (Whiteson and Stone, 2006b; Reeder et al., 2008; Cardamone et al., 2010; Stanley et al., 2016)

Multiagent aggregation task (Silva et al., 2015)

Multiagent integrated navigation & obstacle avoidance (Silva et al., 2015)

Foraging tasks (Silva et al., 2016)

Multiagent Phototaxis task (Silva et al., 2015, 2016)

 

Uniform Moving Average Score (Whiteson and Stone, 2006a)

Utility function (Whiteson and Stone, 2006a; Walsh et al., 2004)

Fitness Value (Zhao et al., 2007; Silva et al., 2015, 2016)

Number of evaluations (Zhao et al., 2007; Silva et al., 2015)

Size of networks (Silva et al., 2015)

Data specific metrics:

-Subjective metric of user's perception (Stanley et al., 2005)

-Users' statistics (Hastings et al., 2009)

-Remaining distance to the target (D'Silva et al., 2005)

-Portion of successful agents (D'Silva et al., 2005)

-Evolved content (Hastings et al., 2009)

-Average lap time in a car simulator (Cardamone et al., 2010)

Controllers' operation time (Silva et al., 2016)

 
Table 9:
Datasets and performance metrics of x-NEAT methods belonging to cluster 2.
DatasetsPerformance metrics

Hybrid NE & BP

 

(NEAT+Q, Online NEAT+Q, L-NEAT, EXACT, Deep HyperNEAT)

 

The mountain car task (Whiteson and Stone, 2006a; Boyan and Moore, 1995)

Server Job Scheduling (Whiteson and Stone, 2006a)

IRIS flowers classification (Chen and Alahakoon, 2006)

Scale Balance Classification (Chen and Alahakoon, 2006)

MNIST handwritten digits (Verbancsics and Harguess, 2015; Desell, 2017a; LeCun, Cortes, et al., 1998)

BCCT200 ship recognition dataset (Verbancsics and Harguess, 2015; Rainey and Stastny, 2011)

 

Uniform Moving Average Score (Whiteson and Stone, 2006a)

Utility function (Whiteson and Stone, 2006a; Walsh et al., 2004)

Accuracy (Chen and Alahakoon, 2006; Verbancsics and Harguess, 2015)

Fitness Value (Verbancsics and Harguess, 2015)

Error Value (Desell, 2017a)

Number of epochs (Desell, 2017a)

Size of networks (Desell, 2017a)

 

Hybrid NE & EC

 

(NEAT+Q, Online NEAT+Q, KO-NEAT, NEAR, PIGEON, FNS-NEATFields, NS-FE-CPPN-NEAT,

MAP-Elites CPPN)

 

Chaser task simulation (Stein et al., 2015)

Sheep task simulation (Stein et al., 2015)

Car task simulation (Stein et al., 2015)

The Mackey–Glass time series (Chatzidimitriou and Mitkas, 2013)

The multiple superimposed oscillator (MSO) (Chatzidimitriou and Mitkas, 2013)

The Lorentz attractor time series (Chatzidimitriou and Mitkas, 2013)

The mountain car task (Chatzidimitriou and Mitkas, 2013; Singh and Sutton, 1996; Whiteson and Stone, 2006a; Boyan and Moore, 1995)

The pole balancing (Chatzidimitriou and Mitkas, 2013; Inden et al., 2013)

Server Job Scheduling (Chatzidimitriou and Mitkas, 2013; Whiteson and Stone, 2006a)

Keepaway Soccer (Zhao et al., 2007; Stone et al., 2005)

Evolving the morphology and the locomotion of soft robots (Methenitis et al., 2015)

Maze navigation (Inden et al., 2013)

Distinction of 4 visual patterns (Inden et al., 2013)

Distinction of 2 visual patterns with variable position (Inden et al., 2013)

Robot locomotion (Tarapore et al., 2016)

 

Similarity factor (Stein and Gonzalez, 2010; Stein et al., 2015)

Uniform Moving Average Score (Whiteson and Stone, 2006a)

Utility function (Whiteson and Stone, 2006a; Walsh et al., 2004)

Fitness Value (Stein et al., 2015; Methenitis et al., 2015; Zhao et al., 2007)

Achieved fitness divided by the number of generations (Inden et al., 2013)

Accumulated reward (Chatzidimitriou and Mitkas, 2013)

Number of evaluations (Chatzidimitriou and Mitkas, 2013; Zhao et al., 2007)

Error value (Chatzidimitriou and Mitkas, 2013)

Global Fitness (Tarapore et al., 2016)

Coverage (Tarapore et al., 2016)

Global reliability (Tarapore et al., 2016)

Precision (Tarapore et al., 2016)

 

Hybrid NE & RL

 

(NEAT+Q, Online NEAT+Q, KO-NEAT, RL-SANE, Online NEAT, Online rtNEAT, NEAR, NEAT-RAC-PGS)

 

The mountain car task (Whiteson and Stone, 2006a; Wright and Gemelli, 2009; Chatzidimitriou and Mitkas, 2013; Peng et al., 2017; Boyan and Moore, 1995)

Server Job Scheduling (Whiteson and Stone, 2006a; Chatzidimitriou and Mitkas, 2013)

Double inverted pendulum (Wright and Gemelli, 2009; Gomez and Miikkulainen, 1999)

The Open Racing Car Simulator (TORCS) (Stanley et al., 2016; Whiteson and Stone, 2006b; Reeder et al., 2008; Cardamone et al., 2010)

Keepaway Soccer (Zhao et al., 2007; Stone et al., 2005)

The Mackey–Glass time series (Chatzidimitriou and Mitkas, 2013)

The multiple superimposed oscillator (MSO) (Chatzidimitriou and Mitkas, 2013)

The Lorentz attractor time series (Chatzidimitriou and Mitkas, 2013)

The pole balancing (Chatzidimitriou and Mitkas, 2013; Peng et al., 2017)

 

Uniform Moving Average Score (Whiteson and Stone, 2006a)

Utility function (Whiteson and Stone, 2006a; Walsh et al., 2004)

Fitness value (Zhao et al., 2007)

Error value (Chatzidimitriou and Mitkas, 2013)

Number of (time) steps (Wright and Gemelli, 2009; Peng et al., 2017)

Number of Evaluations (Chatzidimitriou and Mitkas, 2013; Zhao et al., 2007)

Accumulated reward (Chatzidimitriou and Mitkas, 2013)

Data specific metrics:

-Average lap time in a car simulator (Cardamone et al., 2010)

 

Hybrid NE & UL

 

(NEAT-LSTM-IM, NEAT-FLEX)

 

Sequence classification and recall in T-maze (Rawal and Miikkulainen, 2016)

Protein Data Bank (Grisci and Dorn, 2017)

 

Ratio of successful runs (Rawal and Miikkulainen, 2016)

Accuracy (Grisci and Dorn, 2017)

Size of networks (Grisci and Dorn, 2017)

Task-specific metrics (Grisci and Dorn, 2017)

 
DatasetsPerformance metrics

Hybrid NE & BP

 

(NEAT+Q, Online NEAT+Q, L-NEAT, EXACT, Deep HyperNEAT)

 

The mountain car task (Whiteson and Stone, 2006a; Boyan and Moore, 1995)

Server Job Scheduling (Whiteson and Stone, 2006a)

IRIS flowers classification (Chen and Alahakoon, 2006)

Scale Balance Classification (Chen and Alahakoon, 2006)

MNIST handwritten digits (Verbancsics and Harguess, 2015; Desell, 2017a; LeCun, Cortes, et al., 1998)

BCCT200 ship recognition dataset (Verbancsics and Harguess, 2015; Rainey and Stastny, 2011)

 

Uniform Moving Average Score (Whiteson and Stone, 2006a)

Utility function (Whiteson and Stone, 2006a; Walsh et al., 2004)

Accuracy (Chen and Alahakoon, 2006; Verbancsics and Harguess, 2015)

Fitness Value (Verbancsics and Harguess, 2015)

Error Value (Desell, 2017a)

Number of epochs (Desell, 2017a)

Size of networks (Desell, 2017a)

 

Hybrid NE & EC

 

(NEAT+Q, Online NEAT+Q, KO-NEAT, NEAR, PIGEON, FNS-NEATFields, NS-FE-CPPN-NEAT,

MAP-Elites CPPN)

 

Chaser task simulation (Stein et al., 2015)

Sheep task simulation (Stein et al., 2015)

Car task simulation (Stein et al., 2015)

The Mackey–Glass time series (Chatzidimitriou and Mitkas, 2013)

The multiple superimposed oscillator (MSO) (Chatzidimitriou and Mitkas, 2013)

The Lorentz attractor time series (Chatzidimitriou and Mitkas, 2013)

The mountain car task (Chatzidimitriou and Mitkas, 2013; Singh and Sutton, 1996; Whiteson and Stone, 2006a; Boyan and Moore, 1995)

The pole balancing (Chatzidimitriou and Mitkas, 2013; Inden et al., 2013)

Server Job Scheduling (Chatzidimitriou and Mitkas, 2013; Whiteson and Stone, 2006a)

Keepaway Soccer (Zhao et al., 2007; Stone et al., 2005)

Evolving the morphology and the locomotion of soft robots (Methenitis et al., 2015)

Maze navigation (Inden et al., 2013)

Distinction of 4 visual patterns (Inden et al., 2013)

Distinction of 2 visual patterns with variable position (Inden et al., 2013)

Robot locomotion (Tarapore et al., 2016)

 

Similarity factor (Stein and Gonzalez, 2010; Stein et al., 2015)

Uniform Moving Average Score (Whiteson and Stone, 2006a)

Utility function (Whiteson and Stone, 2006a; Walsh et al., 2004)

Fitness Value (Stein et al., 2015; Methenitis et al., 2015; Zhao et al., 2007)

Achieved fitness divided by the number of generations (Inden et al., 2013)

Accumulated reward (Chatzidimitriou and Mitkas, 2013)

Number of evaluations (Chatzidimitriou and Mitkas, 2013; Zhao et al., 2007)

Error value (Chatzidimitriou and Mitkas, 2013)

Global Fitness (Tarapore et al., 2016)

Coverage (Tarapore et al., 2016)

Global reliability (Tarapore et al., 2016)

Precision (Tarapore et al., 2016)

 

Hybrid NE & RL

 

(NEAT+Q, Online NEAT+Q, KO-NEAT, RL-SANE, Online NEAT, Online rtNEAT, NEAR, NEAT-RAC-PGS)

 

The mountain car task (Whiteson and Stone, 2006a; Wright and Gemelli, 2009; Chatzidimitriou and Mitkas, 2013; Peng et al., 2017; Boyan and Moore, 1995)

Server Job Scheduling (Whiteson and Stone, 2006a; Chatzidimitriou and Mitkas, 2013)

Double inverted pendulum (Wright and Gemelli, 2009; Gomez and Miikkulainen, 1999)

The Open Racing Car Simulator (TORCS) (Stanley et al., 2016; Whiteson and Stone, 2006b; Reeder et al., 2008; Cardamone et al., 2010)

Keepaway Soccer (Zhao et al., 2007; Stone et al., 2005)

The Mackey–Glass time series (Chatzidimitriou and Mitkas, 2013)

The multiple superimposed oscillator (MSO) (Chatzidimitriou and Mitkas, 2013)

The Lorentz attractor time series (Chatzidimitriou and Mitkas, 2013)

The pole balancing (Chatzidimitriou and Mitkas, 2013; Peng et al., 2017)

 

Uniform Moving Average Score (Whiteson and Stone, 2006a)

Utility function (Whiteson and Stone, 2006a; Walsh et al., 2004)

Fitness value (Zhao et al., 2007)

Error value (Chatzidimitriou and Mitkas, 2013)

Number of (time) steps (Wright and Gemelli, 2009; Peng et al., 2017)

Number of Evaluations (Chatzidimitriou and Mitkas, 2013; Zhao et al., 2007)

Accumulated reward (Chatzidimitriou and Mitkas, 2013)

Data specific metrics:

-Average lap time in a car simulator (Cardamone et al., 2010)

 

Hybrid NE & UL

 

(NEAT-LSTM-IM, NEAT-FLEX)

 

Sequence classification and recall in T-maze (Rawal and Miikkulainen, 2016)

Protein Data Bank (Grisci and Dorn, 2017)

 

Ratio of successful runs (Rawal and Miikkulainen, 2016)

Accuracy (Grisci and Dorn, 2017)

Size of networks (Grisci and Dorn, 2017)

Task-specific metrics (Grisci and Dorn, 2017)

 
Table 10:
Datasets and performance metrics of x-NEAT methods belonging to cluster 3.
DatasetsPerformance metrics

ANNs with different types of nodes

 

(CPPN-NEAT, NEAT-CTRNN, TL-CPPN-NEAT, RBF-NEAT, Recurrent CPPN-NEAT, EXACT,

 

SNAP-NEAT, SUPG-HyperNEAT, MAP-Elites CPPN, NEAT-LSTM, NEAT-LSTM-IM, NEAR, NS-FE-CPPN-NEAT, HA-NEAT)

 

Pattern generation (Stanley, 2006)

Pole balancing (Miguel et al., 2008; Kohl and Miikkulainen, 2012; Chatzidimitriou and Mitkas, 2013)

Board game (Bahçeci and Miikkulainen, 2008)

Artificial data with maximal variations (Kohl and Miikkulainen, 2009)

Approximation of the sin(αx) function (Kohl and Miikkulainen, 2009)

The concentric spirals (Kohl and Miikkulainen, 2009, 2012; Potter and Jong, 2000)

The multiplexer (Kohl and Miikkulainen, 2009, 2012)

N-Point classification task (Kohl and Miikkulainen, 2012)

Communication's optimization between evolution of robots' morphology (Auerbach and Bongard, 2011)

MNIST handwritten digits (Desell, 2017a; LeCun, Cortes, et al., 1998)

Half-field Soccer (Kohl and Miikkulainen, 2012; Kalyanakrishnan et al., 2006)

Quadruped Robot Gait control (Auerbach and Bongard, 2011; Smith, 2001)

Sequence classification and recall in T-maze (Rawal and Miikkulainen, 2016)

Server Job Scheduling (Whiteson and Stone, 2006a)

The mountain car task (Chatzidimitriou and Mitkas, 2013; Singh and Sutton, 1996)

The Mackey–Glass time series (Chatzidimitriou and Mitkas, 2013)

The multiple superimposed oscillator (MSO) (Chatzidimitriou and Mitkas, 2013)

The Lorentz attractor time series (Chatzidimitriou and Mitkas, 2013)

Evolving the morphology and the locomotion of soft robots (Methenitis et al., 2015)

Cholesterol Level Indicators (Hagg et al., 2017; Purdie et al., 1992)

Engine Torque and Emissions (Hagg et al., 2017; Gomez and Miikkulainen, 1998)

Wisconsin Breast Cancer Diagnosis Problem (Hagg et al., 2017; Mangasarian and Wolberg, 1990)

Robot locomotion (Tarapore et al., 2016)

 

Fitness value (Auerbach and Bongard, 2011; Methenitis et al., 2015)

Accumulated reward (Chatzidimitriou and Mitkas, 2013)

Number of fitness evaluations (Miguel et al., 2008)

Generalization score (Miguel et al., 2008)

Error value (Hagg et al., 2017)

Generalization metric:

-Ratio of successful runs with different initial conditions (Miguel et al., 2008)

Ratio of successful runs (Rawal and Miikkulainen, 2016)

Size of networks (Miguel et al., 2008; Auerbach and Bongard, 2011; Desell, 2017a; Morse et al., 2013; Hagg et al., 2017)

Game's score (Bahçeci and Miikkulainen, 2008)

Number of generations (Bahçeci and Miikkulainen, 2008)

Number of evaluations (Chatzidimitriou and Mitkas, 2013)

Metric of fracture: variation (Kohl and Miikkulainen, 2009)

Error Value (Kohl and Miikkulainen, 2009; Desell, 2017a; Chatzidimitriou and Mitkas, 2013)

Number of epochs (Desell, 2017a)

Data specific metrics:

-Subjective metric of user's perception (Stanley, 2006)

-Statistics of the evolved morphologies (Auerbach and Bongard, 2011)

-Data specific scores (Kohl and Miikkulainen, 2012)

-Distance traveled by the robot (Morse et al., 2013; Methenitis et al., 2015)

Global Fitness (Tarapore et al., 2016)

Coverage (Tarapore et al., 2016)

Global reliability (Tarapore et al., 2016)

Precision (Tarapore et al., 2016)

 

ANNs by Automatic Substrate Configuration

 

(ES-HyperNEAT, ES-HyperNEAT-LEO, Adaptive ES-HyperNEAT, MSS-HyperNEAT)

 

Retina classification problem (Risi and Stanley, 2012a; Kashtan and Alon, 2005)

Dual Task (navigation and food gathering) (Risi and Stanley, 2012a)

Maze navigation (Risi and Stanley, 2012a, 2012b)

Multiagent maze exploration and group coordination task (Pugh and Stanley, 2013)

 

Fitness Value (Risi and Stanley, 2012a, 2012b; Pugh and Stanley, 2013)

Ratio of successful runs (Risi and Stanley, 2012a, 2012b; Pugh and Stanley, 2013)

Number of generations (Risi and Stanley, 2012a)

Size of networks (Risi and Stanley, 2012a)

 

Modular ANNs

 

(Modular NEAT, HyperNEAT-LEO, MFF-NEAT, ES-HyperNEAT-LEO, MM-NEAT, HyperNEAT-CCT, MB-HyperNEAT)

 

Artificial board game (Reisinger et al., 2004)

Retina classification problem (Verbancsics and Stanley, 2011; Risi and Stanley, 2012a; Huizinga et al., 2014; Kashtan and Alon, 2005)

The Monks problem (Manning and Walsh, 2012; Thrun et al., 1991)

Heart disease classification (Manning and Walsh, 2012; Prechelt and Informatik, 1994)

Mass spectral classification (Manning and Walsh, 2012)

5 independent XOR problems (Huizinga et al., 2014)

2 Hierachically nested XOR problems (Huizinga et al., 2014)

Ms Pac Man (Schrum and Miikkulainen, 2016)

Team patrol (Schrum et al., 2016)

Lone patrol (Schrum et al., 2016)

Dual Task (Schrum et al., 2016)

Two rooms task (Schrum et al., 2016)

 

Modularity metric: number of crosslinks (Reisinger et al., 2004)

Accuracy (Reisinger et al., 2004; Verbancsics and Stanley, 2011)

Fitness Value (Risi and Stanley, 2012a; Schrum and Miikkulainen, 2016; Schrum et al., 2016)

Average Error (Manning and Walsh, 2012)

Post-learning score (Schrum and Miikkulainen, 2016)

Number of generations (Reisinger et al., 2004; Risi and Stanley, 2012a; Manning and Walsh, 2012)

Percentage of time of mostly used module (Schrum and Miikkulainen, 2016)

Ratio of successful runs (Verbancsics and Stanley, 2011; Risi and Stanley, 2012a)

Visual interpretation of crosslinks (Verbancsics and Stanley, 2011)

Size of networks (Risi and Stanley, 2012a)

Q-score (Huizinga et al., 2014; Newman, 2006)

Modularity metrics (Huizinga et al., 2014):

- Modular decomposition

-Number of solved subproblems

Regularity metric (Huizinga et al., 2014)

 

DNNs

 

(Deep HyperNEAT, EXACT)

 

MNIST handwritten digits (Verbancsics and Harguess, 2015; Desell, 2017a; LeCun, Cortes, et al., 1998)

BCCT200 ship recognition dataset (Verbancsics and Harguess, 2015; Rainey and Stastny, 2011)

 

Accuracy (Verbancsics and Harguess, 2015)

Fitness value (Verbancsics and Harguess, 2015)

Error Value (Desell, 2017a)

Number of epochs (Desell, 2017a)

Size of networks (Desell, 2017a)

 

ANNs with Plasticity

 

(MO-NEAT, Adaptive (ES) HyperNEAT, Adaptive HyperNEATv2, Seeded Adaptive HyperNEAT)

 

Communication's optimization between a sensor node and a base station (Haggett and Chu, 2009)

Anomaly detection in time series data (Haggett and Chu, 2009)

T-maze navigation (Risi and Stanley, 2010)

Maze navigation (Risi and Stanley, 2012b)

The Open Racing Car Simulator (TORCS) (Gallego-Durán et al., 2013; Stanley et al., 2016)

Line orientation task (Risi and Stanley, 2014)

 

Fitness Value (Risi and Stanley, 2010, 2012b, 2014)

Game's score (Gallego-Durán et al., 2013)

Number of generations (Risi and Stanley, 2010, 2014)

Computational time (Gallego-Durán et al., 2013)

Percentage of True Positives and False Positives (Haggett and Chu, 2009)

Ratio of successful runs (Risi and Stanley, 2012b)

Data specific metrics:

-Percentage of transmitted packets (Haggett and Chu, 2009)

 

ANNs with Transfer Learning

 

(TL-CPPN-NEAT, online NEAT, online rtNEAT, PLPS-NEAT-TL)

 

Board game (Bahçeci and Miikkulainen, 2008)

The Open Racing Car Simulator (TORCS) (Cardamone et al., 2010; Stanley et al., 2016)

The Mario benchmark (Hardwick-Smith et al., 2017; Karakovskiy and Togelius, 2012)

 

Fitness value (Hardwick-Smith et al., 2017)

Game's score (Bahçeci and Miikkulainen, 2008)

Number of generations (Bahçeci and Miikkulainen, 2008)

Size of networks (Hardwick-Smith et al., 2017)

Data specific metrics:

-Average lap time in a car simulator (Cardamone et al., 2010)

 

Large ANNs

 

(Modular NEAT, HyperNEAT, HyperNEAT-LEO, Multiagent HyperNEAT, Adaptive HyperNEAT,

 

ES-HyperNEAT(-LEO), Adaptive ES-HyperNEAT, Multiagent HyperNEATv2, Switch HybrID,

 

NEATFields, SUPG-HyperNEAT, MSS-HyperNEAT, FNS-NEATFields, HyperNEAT-CCT,

 

Seeded (Adaptive) HyperNEAT, MB-HyperNEAT, τ-HyperNEAT, Adaptive HyperNEATv2)

 

Artificial board game (Reisinger et al., 2004)

Visual Discrimination (Stanley et al., 2009)

Retina classification problem (Verbancsics and Stanley, 2011; Huizinga et al., 2014; Risi and Stanley, 2012a; Kashtan and Alon, 2005)

Multiagent predator-prey task (D'Ambrosio and Stanley, 2008)

T-maze navigation (Risi and Stanley, 2010)

Maze navigation (Risi and Stanley, 2012a, 2012b; Inden et al., 2013)

Multiagent patrolling task (D'Ambrosio et al., 2011)

Target weights problem (Clune et al., 2011)

The Bit Mirroring Problem (Clune et al., 2011)

Quadruped Controller (Clune et al., 2011)

Finding the large square (Inden et al., 2012; Gauci and Stanley, 2007)

Multiagent maze exploration and group coordination task (Pugh and Stanley, 2013)

Distinguishing orientations of shapes and textures (Inden et al., 2012)

Distinguishing orientations of area borders in gray scale images (Inden et al., 2012)

Evolving coordinated reaching movements for segmented arms (Inden et al., 2012)

Quadruped Robot Gait control (Morse et al., 2013; Inden et al., 2013; Silva et al., 2017)

The Open Car Racing Simulator (Gallego-Durán et al., 2013; Stanley et al., 2016)

Distinction of 4 visual patterns (Inden et al., 2013)

Distinction of 2 visual patterns with variable position (Inden et al., 2013)

Pole balancing (Inden et al., 2013)

5 independent XOR problems (Huizinga et al., 2014)

2 Hierachically nested XOR problems (Huizinga et al., 2014)

Line orientation task (Risi and Stanley, 2014)

Team/Lone patrol (Schrum et al., 2016)

Dual Task (Risi and Stanley, 2012a; Schrum et al., 2016)

Two rooms task (Schrum et al., 2016)

 

Accuracy (Reisinger et al., 2004; Verbancsics and Stanley, 2011)

Fitness Value (Risi and Stanley, 2010, 2012a, 2012b; Clune et al., 2011; Pugh and Stanley, 2013; Risi and Stanley, 2014; Schrum et al., 2016; Silva et al., 2017)

Achieved fitness divided by the number of generations (Inden et al., 2013)

Error Value (Stanley et al., 2009)

Game's score (Gallego-Durán et al., 2013)

Number of generations (Reisinger et al., 2004; Risi and Stanley, 2010; Risi and Stanley, 2012a; D'Ambrosio et al., 2011; Risi and Stanley, 2014)

Number of evaluations (Inden et al., 2012)

Computational Time (Gallego-Durán et al., 2013)

Ratio of successful runs (Verbancsics and Stanley, 2011; Risi and Stanley, 2012b; Risi and Stanley, 2012a; Inden et al., 2012; Pugh and Stanley, 2013)

Visual interpretation of crosslinks (Verbancsics and Stanley, 2011)

Success of a task (D'Ambrosio et al., 2011)

Networks' size (Risi and Stanley, 2012a; Morse et al., 2013)

Generalization ability (Inden et al., 2012)

Data specific metric:

-time remaining after the agents have captured all the preys (D'Ambrosio and Stanley, 2008)

-distance traveled by robot (Clune et al., 2011; Morse et al., 2013)

Generated gaits (Silva et al., 2017)

Q-score (Huizinga et al., 2014; Newman, 2006)

Modularity metrics:

-Modular decomposition (Huizinga et al., 2014)

-number of crosslinks (Reisingeret al., 2004)

-Number of solved subproblems (Huizinga et al., 2014)

Regularity metric (Huizinga et al., 2014)

 

ANNs with Memory Capacity

 

(NEAT-CTRNN, NEAR, NEAT-LSTM, NEAT-LSTM-IM, τ-NEAT, τ-HyperNEAT)

 

Pole balancing (Miguel et al., 2008; Chatzidimitriou and Mitkas, 2013)

Sequence classification and recall in T-maze (Rawal and Miikkulainen, 2016)

The Mackey–Glass time series (Chatzidimitriou and Mitkas, 2013; Caamaño et al., 2016)

The multiple superimposed oscillator (MSO) (Chatzidimitriou and Mitkas, 2013)

The Lorentz attractor time series (Chatzidimitriou and Mitkas, 2013)

The mountain car task (Chatzidimitriou and Mitkas, 2013; Singh and Sutton, 1996)

Server Job Scheduling (Chatzidimitriou and Mitkas, 2013)

Monthly CO2 concentrations (Caamaño et al., 2016; Thoning et al., 1989)

Monthly number of international airline passengers (Caamaño et al., 2016; Box et al., 2015)

Safe Crossing robot task (Caamaño et al., 2016)

Mimicking motion robot task (Caamaño et al., 2016)

Quadruped Gait control (Silva et al., 2017)

 

Fitness value (Silva et al., 2017)

Accumulated reward (Chatzidimitriou and Mitkas, 2013)

Error value (Chatzidimitriou and Mitkas, 2013; Caamaño et al., 2016)

Number of fitness evaluations (Miguel et al., 2008; Chatzidimitriou and Mitkas, 2013)

Generalization score (Miguel et al., 2008):

Generalization metric:

-Ratio of successful runs with different initial conditions (Miguel et al., 2008)

Ratio of successful runs (Rawal and Miikkulainen, 2016)

Size of networks (Miguel et al., 2008)

Generated gaits (Silva et al., 2017)

 
DatasetsPerformance metrics

ANNs with different types of nodes

 

(CPPN-NEAT, NEAT-CTRNN, TL-CPPN-NEAT, RBF-NEAT, Recurrent CPPN-NEAT, EXACT,

 

SNAP-NEAT, SUPG-HyperNEAT, MAP-Elites CPPN, NEAT-LSTM, NEAT-LSTM-IM, NEAR, NS-FE-CPPN-NEAT, HA-NEAT)

 

Pattern generation (Stanley, 2006)

Pole balancing (Miguel et al., 2008; Kohl and Miikkulainen, 2012; Chatzidimitriou and Mitkas, 2013)

Board game (Bahçeci and Miikkulainen, 2008)

Artificial data with maximal variations (Kohl and Miikkulainen, 2009)

Approximation of the sin(αx) function (Kohl and Miikkulainen, 2009)

The concentric spirals (Kohl and Miikkulainen, 2009, 2012; Potter and Jong, 2000)

The multiplexer (Kohl and Miikkulainen, 2009, 2012)

N-Point classification task (Kohl and Miikkulainen, 2012)

Communication's optimization between evolution of robots' morphology (Auerbach and Bongard, 2011)

MNIST handwritten digits (Desell, 2017a; LeCun, Cortes, et al., 1998)

Half-field Soccer (Kohl and Miikkulainen, 2012; Kalyanakrishnan et al., 2006)

Quadruped Robot Gait control (Auerbach and Bongard, 2011; Smith, 2001)

Sequence classification and recall in T-maze (Rawal and Miikkulainen, 2016)

Server Job Scheduling (Whiteson and Stone, 2006a)

The mountain car task (Chatzidimitriou and Mitkas, 2013; Singh and Sutton, 1996)

The Mackey–Glass time series (Chatzidimitriou and Mitkas, 2013)

The multiple superimposed oscillator (MSO) (Chatzidimitriou and Mitkas, 2013)

The Lorentz attractor time series (Chatzidimitriou and Mitkas, 2013)

Evolving the morphology and the locomotion of soft robots (Methenitis et al., 2015)

Cholesterol Level Indicators (Hagg et al., 2017; Purdie et al., 1992)

Engine Torque and Emissions (Hagg et al., 2017; Gomez and Miikkulainen, 1998)

Wisconsin Breast Cancer Diagnosis Problem (Hagg et al., 2017; Mangasarian and Wolberg, 1990)

Robot locomotion (Tarapore et al., 2016)

 

Fitness value (Auerbach and Bongard, 2011; Methenitis et al., 2015)

Accumulated reward (Chatzidimitriou and Mitkas, 2013)

Number of fitness evaluations (Miguel et al., 2008)

Generalization score (Miguel et al., 2008)

Error value (Hagg et al., 2017)

Generalization metric:

-Ratio of successful runs with different initial conditions (Miguel et al., 2008)

Ratio of successful runs (Rawal and Miikkulainen, 2016)

Size of networks (Miguel et al., 2008; Auerbach and Bongard, 2011; Desell, 2017a; Morse et al., 2013; Hagg et al., 2017)

Game's score (Bahçeci and Miikkulainen, 2008)

Number of generations (Bahçeci and Miikkulainen, 2008)

Number of evaluations (Chatzidimitriou and Mitkas, 2013)

Metric of fracture: variation (Kohl and Miikkulainen, 2009)

Error Value (Kohl and Miikkulainen, 2009; Desell, 2017a; Chatzidimitriou and Mitkas, 2013)

Number of epochs (Desell, 2017a)

Data specific metrics:

-Subjective metric of user's perception (Stanley, 2006)

-Statistics of the evolved morphologies (Auerbach and Bongard, 2011)

-Data specific scores (Kohl and Miikkulainen, 2012)

-Distance traveled by the robot (Morse et al., 2013; Methenitis et al., 2015)

Global Fitness (Tarapore et al., 2016)

Coverage (Tarapore et al., 2016)

Global reliability (Tarapore et al., 2016)

Precision (Tarapore et al., 2016)

 

ANNs by Automatic Substrate Configuration

 

(ES-HyperNEAT, ES-HyperNEAT-LEO, Adaptive ES-HyperNEAT, MSS-HyperNEAT)

 

Retina classification problem (Risi and Stanley, 2012a; Kashtan and Alon, 2005)

Dual Task (navigation and food gathering) (Risi and Stanley, 2012a)

Maze navigation (Risi and Stanley, 2012a, 2012b)

Multiagent maze exploration and group coordination task (Pugh and Stanley, 2013)

 

Fitness Value (Risi and Stanley, 2012a, 2012b; Pugh and Stanley, 2013)

Ratio of successful runs (Risi and Stanley, 2012a, 2012b; Pugh and Stanley, 2013)

Number of generations (Risi and Stanley, 2012a)

Size of networks (Risi and Stanley, 2012a)

 

Modular ANNs

 

(Modular NEAT, HyperNEAT-LEO, MFF-NEAT, ES-HyperNEAT-LEO, MM-NEAT, HyperNEAT-CCT, MB-HyperNEAT)

 

Artificial board game (Reisinger et al., 2004)

Retina classification problem (Verbancsics and Stanley, 2011; Risi and Stanley, 2012a; Huizinga et al., 2014; Kashtan and Alon, 2005)

The Monks problem (Manning and Walsh, 2012; Thrun et al., 1991)

Heart disease classification (Manning and Walsh, 2012; Prechelt and Informatik, 1994)

Mass spectral classification (Manning and Walsh, 2012)

5 independent XOR problems (Huizinga et al., 2014)

2 Hierachically nested XOR problems (Huizinga et al., 2014)

Ms Pac Man (Schrum and Miikkulainen, 2016)

Team patrol (Schrum et al., 2016)

Lone patrol (Schrum et al., 2016)

Dual Task (Schrum et al., 2016)

Two rooms task (Schrum et al., 2016)

 

Modularity metric: number of crosslinks (Reisinger et al., 2004)

Accuracy (Reisinger et al., 2004; Verbancsics and Stanley, 2011)

Fitness Value (Risi and Stanley, 2012a; Schrum and Miikkulainen, 2016; Schrum et al., 2016)

Average Error (Manning and Walsh, 2012)

Post-learning score (Schrum and Miikkulainen, 2016)

Number of generations (Reisinger et al., 2004; Risi and Stanley, 2012a; Manning and Walsh, 2012)

Percentage of time of mostly used module (Schrum and Miikkulainen, 2016)

Ratio of successful runs (Verbancsics and Stanley, 2011; Risi and Stanley, 2012a)

Visual interpretation of crosslinks (Verbancsics and Stanley, 2011)

Size of networks (Risi and Stanley, 2012a)

Q-score (Huizinga et al., 2014; Newman, 2006)

Modularity metrics (Huizinga et al., 2014):

- Modular decomposition

-Number of solved subproblems

Regularity metric (Huizinga et al., 2014)

 

DNNs

 

(Deep HyperNEAT, EXACT)

 

MNIST handwritten digits (Verbancsics and Harguess, 2015; Desell, 2017a; LeCun, Cortes, et al., 1998)

BCCT200 ship recognition dataset (Verbancsics and Harguess, 2015; Rainey and Stastny, 2011)

 

Accuracy (Verbancsics and Harguess, 2015)

Fitness value (Verbancsics and Harguess, 2015)

Error Value (Desell, 2017a)

Number of epochs (Desell, 2017a)

Size of networks (Desell, 2017a)

 

ANNs with Plasticity

 

(MO-NEAT, Adaptive (ES) HyperNEAT, Adaptive HyperNEATv2, Seeded Adaptive HyperNEAT)

 

Communication's optimization between a sensor node and a base station (Haggett and Chu, 2009)

Anomaly detection in time series data (Haggett and Chu, 2009)

T-maze navigation (Risi and Stanley, 2010)

Maze navigation (Risi and Stanley, 2012b)

The Open Racing Car Simulator (TORCS) (Gallego-Durán et al., 2013; Stanley et al., 2016)

Line orientation task (Risi and Stanley, 2014)

 

Fitness Value (Risi and Stanley, 2010, 2012b, 2014)

Game's score (Gallego-Durán et al., 2013)

Number of generations (Risi and Stanley, 2010, 2014)

Computational time (Gallego-Durán et al., 2013)

Percentage of True Positives and False Positives (Haggett and Chu, 2009)

Ratio of successful runs (Risi and Stanley, 2012b)

Data specific metrics:

-Percentage of transmitted packets (Haggett and Chu, 2009)

 

ANNs with Transfer Learning

 

(TL-CPPN-NEAT, online NEAT, online rtNEAT, PLPS-NEAT-TL)

 

Board game (Bahçeci and Miikkulainen, 2008)

The Open Racing Car Simulator (TORCS) (Cardamone et al., 2010; Stanley et al., 2016)

The Mario benchmark (Hardwick-Smith et al., 2017; Karakovskiy and Togelius, 2012)

 

Fitness value (Hardwick-Smith et al., 2017)

Game's score (Bahçeci and Miikkulainen, 2008)

Number of generations (Bahçeci and Miikkulainen, 2008)

Size of networks (Hardwick-Smith et al., 2017)

Data specific metrics:

-Average lap time in a car simulator (Cardamone et al., 2010)

 

Large ANNs

 

(Modular NEAT, HyperNEAT, HyperNEAT-LEO, Multiagent HyperNEAT, Adaptive HyperNEAT,

 

ES-HyperNEAT(-LEO), Adaptive ES-HyperNEAT, Multiagent HyperNEATv2, Switch HybrID,

 

NEATFields, SUPG-HyperNEAT, MSS-HyperNEAT, FNS-NEATFields, HyperNEAT-CCT,

 

Seeded (Adaptive) HyperNEAT, MB-HyperNEAT, τ-HyperNEAT, Adaptive HyperNEATv2)

 

Artificial board game (Reisinger et al., 2004)

Visual Discrimination (Stanley et al., 2009)

Retina classification problem (Verbancsics and Stanley, 2011; Huizinga et al., 2014; Risi and Stanley, 2012a; Kashtan and Alon, 2005)

Multiagent predator-prey task (D'Ambrosio and Stanley, 2008)

T-maze navigation (Risi and Stanley, 2010)

Maze navigation (Risi and Stanley, 2012a, 2012b; Inden et al., 2013)

Multiagent patrolling task (D'Ambrosio et al., 2011)

Target weights problem (Clune et al., 2011)

The Bit Mirroring Problem (Clune et al., 2011)

Quadruped Controller (Clune et al., 2011)

Finding the large square (Inden et al., 2012; Gauci and Stanley, 2007)

Multiagent maze exploration and group coordination task (Pugh and Stanley, 2013)

Distinguishing orientations of shapes and textures (Inden et al., 2012)

Distinguishing orientations of area borders in gray scale images (Inden et al., 2012)

Evolving coordinated reaching movements for segmented arms (Inden et al., 2012)

Quadruped Robot Gait control (Morse et al., 2013; Inden et al., 2013; Silva et al., 2017)

The Open Car Racing Simulator (Gallego-Durán et al., 2013; Stanley et al., 2016)

Distinction of 4 visual patterns (Inden et al., 2013)

Distinction of 2 visual patterns with variable position (Inden et al., 2013)

Pole balancing (Inden et al., 2013)

5 independent XOR problems (Huizinga et al., 2014)

2 Hierachically nested XOR problems (Huizinga et al., 2014)

Line orientation task (Risi and Stanley, 2014)

Team/Lone patrol (Schrum et al., 2016)

Dual Task (Risi and Stanley, 2012a; Schrum et al., 2016)

Two rooms task (Schrum et al., 2016)

 

Accuracy (Reisinger et al., 2004; Verbancsics and Stanley, 2011)

Fitness Value (Risi and Stanley, 2010, 2012a, 2012b; Clune et al., 2011; Pugh and Stanley, 2013; Risi and Stanley, 2014; Schrum et al., 2016; Silva et al., 2017)

Achieved fitness divided by the number of generations (Inden et al., 2013)

Error Value (Stanley et al., 2009)

Game's score (Gallego-Durán et al., 2013)

Number of generations (Reisinger et al., 2004; Risi and Stanley, 2010; Risi and Stanley, 2012a; D'Ambrosio et al., 2011; Risi and Stanley, 2014)

Number of evaluations (Inden et al., 2012)

Computational Time (Gallego-Durán et al., 2013)

Ratio of successful runs (Verbancsics and Stanley, 2011; Risi and Stanley, 2012b; Risi and Stanley, 2012a; Inden et al., 2012; Pugh and Stanley, 2013)

Visual interpretation of crosslinks (Verbancsics and Stanley, 2011)

Success of a task (D'Ambrosio et al., 2011)

Networks' size (Risi and Stanley, 2012a; Morse et al., 2013)

Generalization ability (Inden et al., 2012)

Data specific metric:

-time remaining after the agents have captured all the preys (D'Ambrosio and Stanley, 2008)

-distance traveled by robot (Clune et al., 2011; Morse et al., 2013)

Generated gaits (Silva et al., 2017)

Q-score (Huizinga et al., 2014; Newman, 2006)

Modularity metrics:

-Modular decomposition (Huizinga et al., 2014)

-number of crosslinks (Reisingeret al., 2004)

-Number of solved subproblems (Huizinga et al., 2014)

Regularity metric (Huizinga et al., 2014)

 

ANNs with Memory Capacity

 

(NEAT-CTRNN, NEAR, NEAT-LSTM, NEAT-LSTM-IM, τ-NEAT, τ-HyperNEAT)

 

Pole balancing (Miguel et al., 2008; Chatzidimitriou and Mitkas, 2013)

Sequence classification and recall in T-maze (Rawal and Miikkulainen, 2016)

The Mackey–Glass time series (Chatzidimitriou and Mitkas, 2013; Caamaño et al., 2016)

The multiple superimposed oscillator (MSO) (Chatzidimitriou and Mitkas, 2013)

The Lorentz attractor time series (Chatzidimitriou and Mitkas, 2013)

The mountain car task (Chatzidimitriou and Mitkas, 2013; Singh and Sutton, 1996)

Server Job Scheduling (Chatzidimitriou and Mitkas, 2013)

Monthly CO2 concentrations (Caamaño et al., 2016; Thoning et al., 1989)

Monthly number of international airline passengers (Caamaño et al., 2016; Box et al., 2015)

Safe Crossing robot task (Caamaño et al., 2016)

Mimicking motion robot task (Caamaño et al., 2016)

Quadruped Gait control (Silva et al., 2017)

 

Fitness value (Silva et al., 2017)

Accumulated reward (Chatzidimitriou and Mitkas, 2013)

Error value (Chatzidimitriou and Mitkas, 2013; Caamaño et al., 2016)

Number of fitness evaluations (Miguel et al., 2008; Chatzidimitriou and Mitkas, 2013)

Generalization score (Miguel et al., 2008):

Generalization metric:

-Ratio of successful runs with different initial conditions (Miguel et al., 2008)

Ratio of successful runs (Rawal and Miikkulainen, 2016)

Size of networks (Miguel et al., 2008)

Generated gaits (Silva et al., 2017)

 

Close Modal

or Create an Account

Close Modal
Close Modal