Convolutional Neural Networks (CNNs) have demonstrated their superiority in image classification, and evolutionary computation (EC) methods have recently been surging to automatically design the architectures of CNNs to save the tedious work of manually designing CNNs. In this paper, a new hybrid differential evolution (DE) algorithm with a newly added crossover operator is proposed to evolve the architectures of CNNs of any lengths, which is named DECNN. There are three new ideas in the proposed DECNN method. Firstly, an existing effective encoding scheme is refined to cater for variable-length CNN architectures; Secondly, the new mutation and crossover operators are developed for variable-length DE to optimise the hyperparameters of CNNs; Finally, the new second crossover is introduced to evolve the depth of the CNN architectures. The proposed algorithm is tested on six widely-used benchmark datasets and the results are compared to 12 state-of-the-art methods, which shows the proposed method is vigorously competitive to the state-of-the-art algorithms. Furthermore, the proposed method is also compared with a method using particle swarm optimisation with a similar encoding strategy named IPPSO, and the proposed DECNN outperforms IPPSO in terms of the accuracy.

Over the past several weeks I have been tinkering with an "art generator AI". It's very much a work in progress but people seem to have opinions about this sort of thing so I thought I'd share what I've done, or what it has done, or both, depending on your point of view. I'm sure we've all been part of or heard some version of the nature of art or "is this art?" debate. Algorithmic and AI generated art seems to attract particular questioning. If the art is generated then is it my art or the software's art?

Princeton Univ, $29.95 (288p) ISBN 978-0-691-18351-0 The Greeks thought of everything, including sci-fi tropes such as androids and artificial intelligence, according to this lively study of mythology and technology. Stanford classicist Mayor (The Amazons) surveys myths from ancient Greece (with excursions to India and China) about bio-techne, life crafted by artifice. She finds a trove of them, including those of the bronze warrior-robot Talos, who patrolled Crete, hurling boulders at ships and roasting soldiers alive; statues by the legendary engineer Daedalus, so lifelike that they had to be tethered to stay put; and marvels by the blacksmith god Hephaestus, including automated rolling tripods that served Olympian feasts and talking robot servants to help at his forge. Taking a more organic approach, the witch Medea, after defeating Talos with sweet talk and trickery, invented herbal drugs to reverse aging. Mayor also looks at real-life automata in ancient Alexandria--mechanical beasts and people that moved, vocalized, and dispensed milk to bemused onlookers.

Deep neuroevolution, that is evolutionary policy search methods based on deep neural networks, have recently emerged as a competitor to deep reinforcement learning algorithms due to their better parallelization capabilities. However, these methods still suffer from a far worse sample efficiency. In this paper we investigate whether a mechanism known as "importance mixing" can significantly improve their sample efficiency. We provide a didactic presentation of importance mixing and we explain how it can be extended to reuse more samples. Then, from an empirical comparison based on a simple benchmark, we show that, though it actually provides better sample efficiency, it is still far from the sample efficiency of deep reinforcement learning, though it is more stable.

The tutorial starts by presenting the equation that we are going to implement.

The time series classification literature has expanded rapidly over the last decade, with many new classification approaches published each year. The research focus has mostly been on improving the accuracy and efficiency of classifiers, while their interpretability has been somewhat neglected. Classifier interpretability has become a critical constraint for many application domains and the introduction of the 'right to explanation' GDPR EU legislation in May 2018 is likely to further emphasize the importance of explainable learning algorithms. In this work we analyse the state-of-the-art for time series classification, and propose new algorithms that aim to maintain the classifier accuracy and efficiency, but keep interpretability as a key design constraint. We present new time series classification algorithms that advance the state-of-the-art by implementing the following three key ideas: (1) Multiple resolutions of symbolic approximations: we combine symbolic representations obtained using different parameters; (2) Multiple domain representations: we combine symbolic approximations in time (e.g., SAX) and frequency (e.g., SFA) domains; (3) Efficient navigation of a huge symbolic-words space: we adapt a symbolic sequence classifier named SEQL, to make it work with multiple domain representations (e.g., SAX-SEQL, SFA-SEQL), and use its greedy feature selection strategy to effectively filter the best features for each representation. We show that a multi-resolution multi-domain linear classifier, SAX-SFA-SEQL, achieves a similar accuracy to the state-of-the-art COTE ensemble, and to a recent deep learning method (FCN), but uses a fraction of the time required by either COTE or FCN. We discuss the accuracy, efficiency and interpretability of our proposed algorithms. To further analyse the interpretability aspect of our classifiers, we present a case study on an ecology benchmark.

With advances in artificial intelligence impacting every industry from healthcare to retail, it's no wonder people are scared. After all, these pesky machines can already perform a great many tasks better than us humans and it's only going to get worse. I'm not just talking about replacing mindless busywork like sorting mail and processing tax returns – I'm talking about AI systems taking on complex jobs like forecasting financial markets, diagnosing medical patients, even making optimized hiring decisions, and doing it all better than highly trained humans. Consider the field of radiology. To become a practicing radiologist in the US, an aspiring doctor must devote 4 years to undergraduate education, another 4 years to medical school and a final 4 years to a radiology residency program.

Fuzzy clustering methods identify naturally occurring clusters in a dataset, where the extent to which different clusters are overlapped can differ. Most methods have a parameter to fix the level of fuzziness. However, the appropriate level of fuzziness depends on the application at hand. This paper presents Entropy $c$-Means (ECM), a method of fuzzy clustering that simultaneously optimizes two contradictory objective functions, resulting in the creation of fuzzy clusters with different levels of fuzziness. This allows ECM to identify clusters with different degrees of overlap. ECM optimizes the two objective functions using two multi-objective optimization methods, Non-dominated Sorting Genetic Algorithm II (NSGA-II), and Multiobjective Evolutionary Algorithm based on Decomposition (MOEA/D). We also propose a method to select a suitable trade-off clustering from the Pareto front. Experiments on challenging synthetic datasets as well as real-world datasets show that ECM leads to better cluster detection compared to the conventional fuzzy clustering methods as well as previously used multi-objective methods for fuzzy clustering.

Evolving Floor Plans is an experimental research project exploring speculative, optimized floor plan layouts. The rooms and expected flow of people are given to a genetic algorithm which attempts to optimize the layout to minimize walking time, the use of hallways, etc. The creative goal is to approach floor plan design solely from the perspective of optimization and without regard for convention, constructability, etc. The research goal is to see how a combination of explicit, implicit and emergent methods allow floor plans of high complexity to evolve. The floorplan is'grown' from its genetic encoding using indirect methods such as graph contraction and emergent ones such as growing hallways using an ant-colony inspired algorithm.

We consider the problem of mining signal temporal logical requirements from a dataset of regular (good) and anomalous (bad) trajectories of a dynamical system. We assume the training set to be labeled by human experts and that we have access only to a limited amount of data, typically noisy. We provide a systematic approach to synthesize both the syntactical structure and the parameters of the temporal logic formula using a two-steps procedure: first, we leverage a novel evolutionary algorithm for learning the structure of the formula; second, we perform the parameter synthesis operating on the statistical emulation of the average robustness for a candidate formula w.r.t. its parameters. We compare our results with our previous work [{BufoBSBLB14] and with a recently proposed decision-tree [bombara_decision_2016] based method. We present experimental results on two case studies: an anomalous trajectory detection problem of a naval surveillance system and the characterization of an Ineffective Respiratory effort, showing the usefulness of our work.