15

Evolutionary algorithms are a family of optimization algorithms based on the principle of Darwinian natural selection. As part of natural selection, a given environment has a population of individuals that compete for survival and reproduction. The ability of each individual to achieve these goals determines their chance to have children, in other words, to ...


12

Section 4.2 of "Essentials of Metaheuristics" has a wealth of information on alternative ways of encoding graph structures via Genetic Algorithms. With particular regard to evolving ANNs, I would personally not be inclined to implement this sort of thing 'from scratch': The field of neuroevolution has been around for some time, and the implementation some ...


8

"Trap" functions were introduced as a way to discuss how GAs behave on functions where sampling most of the search space would provide pressure for the algorithm to move in the wrong direction (wrong in the sense of away from the global optimum). For example, consider a four-bit function f(x) such that f(0000) = 5 f(0001) = 1 f(0010) = 1 f(0011) = 2 f(0100)...


8

A genetic algorithm is an algorithm that randomly generates a number of attempted solutions for a problem. This set of attempted solutions is called the "population". It then tries to see how well these solutions solve the problem, using a given fitness function. The attempted solutions with the best fitness value are used to generate a new population. ...


7

As explained in an answer to this AI SE question, GAs are 'satisficers' rather than 'optimizers' and tend not to explore 'outlying' regions of the search space. Rather, the population tends to cluster in regions that are 'fairly good' according to the fitness function. In contrast, I believe the thinking is that novelty affords a kind of dynamic fitness, ...


7

Over the last few years, evolutionary computation research has shown increasing interest in including some aspect of epigenetics. For example: A 2008 paper by Tanev and Yuta Work from Lee Spector's genetic programming group A recent paper by Ricalde and Banzhaf


7

Unlike backpropagation, evolutionary algorithms do not require the objective function to be differential with respect to the parameters you aim to optimize. As a result, you can optimize "more things" in the network, such as activation functions or number of layers, which wouldn't be possible in the standard backpropagation. Another advantage is that by ...


7

Mutation is usually defined to be a global operator, i.e. iterated mutation is (eventually) capable of reaching every point in the vector space defined by the geneome. In that sense, mutation alone is certainly 'enough'. Regarding the motivation for crossover - from Essentials of Metaheuristics, p42: Crossover was originally based on the premise that ...


6

Further to Franck's answer, there may be better optima (even global optima) that exist in the opposite direction to the gradient (which may be in the direction of some local optima). Evolutionary algorithms have scope to search the surrounding area, while backpropagation will always move in the direction of the gradient. With no guarantee (due to their ...


6

Novelty search selects for "novel behavior", by some domain-dependent definition of novelty. For example, novelty in a Maze-solving domain might be "difference of route explored". Eventually, networks that take every possible route through the maze will be found, and you could then select the fastest. This would work far better than a naive "objective", like ...


6

There are a number of good answers here explaining what genetic algorithms are, and giving example applications. I'm adding some general purpose advice on what they are good for, but also cases where you should NOT use them. If my tone seems harsh, it is because using GAs in any of the cases in the inappropriate section below will lead to your paper being ...


6

This answer requests a practical example of how one might be used, which I will attempt to provide in addition to the other answers. They seem to due a very good job of explaining what a genetic algorithm is. So, this will give an example. Let's say you have a neural network (although they are not the only application of it), which, from some given inputs, ...


6

Take a robot that we want to be able to move from the bottom right corner to the top left corner of a 4x4 matrix full of random holes it should avoid. With holes represented by 1s, it could look something like: exit \/ [0,0,0,1] [0,1,1,0] [0,1,1,1] [0,0,0,0] /\ enter As we want it to get to an exit from a start, we have a natural fitness ...


6

I don't think there is a single standard word or phrase that covers just this concept. Perhaps recursive self-improvement matches the idea concisely - but that is not specific AI jargon. Very little is understood about what strength this effect can have or what the limits are. Will 10 generations of self-improvement lead to a machine that is 10% better, 10 ...


6

Evolutionary algorithms (EAs) are a family of algorithms inspired by the biological evolution that can be used to solve (constrained or not) optimization problems where the function that needs to be optimized does not necessarily need to be differentiable (or satisfy any strong constraint). In EAs, you typically only need to define an encoding of the ...


5

A feed forward neural network without hidden nodes can only find linear decision boundaries. However, most of the time you need non-linear decision boundaries. Hence you need hidden nodes with a non-linear activation function. The more hidden nodes you have, the more data you need to find good parameters, but the more complex decision boundaries you can find....


5

There have been extensive studies within evolutionary computation in the area of island models and niching for doing exactly this. The advantages of this approach include greater population diversity (which is particularly useful when the problem is multiobjective) and the potential for concurrent execution of each separate population. See also the answers ...


5

As observed in another answer, all you need to apply Genetic Algorithms (GAs) is to represent a potential solution to your problem in a form that is subject to crossover and mutation. Ideally, the fitness function will provide some kind of smooth feedback about the quality of a solution, rather than simply being a 'Needle in a Haystack'. Here are some ...


5

The core question to whether or not an AI is adaptable or not is whether or not it supports online learning. That doesn't mean using the Internet to learn things; that means continuing to accept training data during the functioning of the system. This is (mostly) independent of the underlying architecture; in evolutionary approaches one can continue to ...


5

Crossover allows to combine two parents (vs. mutation, which only uses one parent). In some cases, it is useful (e.g., if you train a FPS bot, if one parent is good at shooting and another parent is good at moving, it makes sense to combine them). In some other cases, it is not.


5

The mutation is an operation that is applied to a single individual in the population. It can e.g. introduce some noise in the chromosome. For example, if the chromosomes are binary, a mutation may simply be the flip of a bit (or gene). The crossover is an operation which takes as input two individuals (often called the "parents") and somehow ...


5

Yes, this is an active area of research as we speak. Both using classic algorithms (decision trees, random forests, Bayesian ensembles) as well as neural networks. This can also be done via evolutionary algorithms. I have personally used them for hyperparameter tuning in a few cases where squeezing out a couple of extra points of accuracy was key. This is ...


4

When thinking about crossover its important to think about the fitness landscape. Consider a hypothetical scenario where we are applying a genetic algorithm to find a solution that performs well at 2 tasks. This could be from Franck's example (moving and shooting) for an AI, or perhaps it could be predicted 2 outputs in a genetic machine learning scenario, ...


4

If the system claims that a piece of code has violated standards, then to be useful to the programmer, it really needs to provide more information than just a 'yes/no' classifier: you need some form of explanation about why it is claimed to be wrong. Clearly ANNs aren't much use for that. If I were tackling such a problem (and my suspicion is that a lot of ...


4

Using evolutionary algorithms to evolve neural networks is called neuroevolution. Some neuroevolution algorithms optimize only the weights of a neural network with fixed topology. That sounds not like what you want. Other neuroevolution algorithms optimize both the weights and topology of a neural net. These kinds of algorithms seem more appropriate for ...


4

NEAT has a constant number of organisms in its population, which prevents overpopulation from happening. The process of mating includes the following steps. The worst networks from every species are removed. All species receive a number of offsprings that they can have. This is calculated by an adjusted neural network fitness. Offsprings for species are ...


4

It is actually the other way around: connection IDs is what is debated! Nodes always have innovation IDs (in the image, it is just their identifying number). Node IDs are sufficient to identify connections. If a connection links nodes 3 and 6, then it is the same as another connection linking nodes 3 and 6: no need for an extra ID. So why the extra ...


4

A human analogy can help you here (a variance). Initialize all the agents with an initial value $x$; we will call this energyUnits. I Will talk later more about this. Now, add some value, as an incentive, whenever the agent eats good food, to the energyUnits. You need to add a function that will keep decrementing the value of the agent's energyUnits, as ...


4

Biological and artificial evolution work around pretty much the same principles. Fitness and selection: In biology, the fittest organisms in an ecosystem are more likely to survive long enough to reproduce, passing on their genes in the process. In artificial evolution, our organisms are in fact solutions to our problem, which can be evaluated to determine ...


4

The mutation operation is (usually) needed to introduce new genes not found in the population. For example, suppose that you have 4 possible genes $A$, $B$, $C$, and $D$, and that your chromosomes have a non-binary encoding. In that case, if no member of your population has the gene $D$, then no amount of crossover operations will result in the introduction ...


Only top voted, non community-wiki answers of a minimum length are eligible