Computer Science

With neural networks having demonstrated their versatility and benefits, the need for their optimal performance is as prevalent as ever. A defining characteristic, hyperparameters, can greatly affect its performance. Thus engineers go through a process, tuning, to identify and implement optimal hyperparameters. That being said, excess amounts of manual effort are required for tuning network architectures, training configurations, and preprocessing settings such as Byte Pair Encoding (BPE). In this study, we propose an automatic tuning method modeled after Darwin's Survival of the Fittest Theory via a Genetic Algorithm (GA). Research results show that the proposed method, a GA, outperforms a random selection of hyperparameters.

Modern technological advancements constantly push forward the human-machine interaction. Evolutionary Algorithms (EA) are an machine learning (ML) subclass inspired by the process of natural selection - Survival of the Fittest, as stated by the Darwinian Theory of Evolution. The most notable algorithm in that class is the Genetic Algorithm (GA) - a powerful heuristic tool which enables the generation of a high-quality solutions to optimization problems. In recent decades the algorithm underwent remarkable improvement, which adapted it into a wide range of engineering problems, by heuristically searching for the optimal solution. Despite being well-defined, many engineering problems may suffer from heavy analytical entanglement when approaching the derivation process, as required in classic optimization methods. Therefore, the main motivation here, is to work around that obstacle. In this piece of work, I would like to harness the GA capabilities to examine optimality with respect to a unique combustion problem, in a way that was never performed before. To be more precise, I would like to utilize it to answer the question : What form of an initial droplet size distribution (iDSD) will guarantee an optimal flame ? To answer this question, I will first provide a general introduction to the GA method, then develop the combustion model, and eventually merge both into an optimization problem.

The range of applications of traditional optimization methods are limited by the features of the object variables, and of both the objective and the constraint functions. In contrast, population-based algorithms whose optimization capabilities are emergent properties, such as evolutionary algorithms and particle swarm optimization, present almost no restriction on those features and can handle different optimization problems with few or no adaptations. Their main drawbacks consist of their comparatively higher computational cost and difficulty in handling equality constraints. The particle swarm optimization method is sometimes viewed as an evolutionary algorithm because of their many similarities, despite not being inspired by the same metaphor: they evolve a population of individuals taking into account previous experiences and using stochastic operators to introduce new responses. The advantages of evolutionary algorithms with respect to traditional methods have been greatly discussed in the literature for decades. While the particle swarm optimizers share such advantages, their main desirable features when compared to evolutionary algorithms are their lower computational cost and easier implementation, involving no operator design and few parameters to be tuned. However, even slight modifications of these parameters greatly influence the dynamics of the swarm. This paper deals with the effect of the settings of the parameters of the particles' velocity update equation on the behaviour of the system.

Problem decomposition plays a vital role when applying cooperative coevolution (CC) to large scale global optimization problems. However, most learning-based decomposition algorithms either only apply to additively separable problems or face the issue of false separability detections. Directing against these limitations, this study proposes a novel decomposition algorithm called surrogate-assisted variable grouping (SVG). SVG first designs a general-separability-oriented detection criterion according to whether the optimum of a variable changes with other variables. This criterion is consistent with the separability definition and thus endows SVG with broad applicability and high accuracy. To reduce the fitness evaluation requirement, SVG seeks the optimum of a variable with the help of a surrogate model rather than the original expensive high-dimensional model. Moreover, it converts the variable grouping process into a dynamic-binary-tree search one, which facilitates reutilizing historical separability detection information and thus reducing detection times. To evaluate the performance of SVG, a suite of benchmark functions with up to 2000 dimensions, including additively and non-additively separable ones, were designed. Experimental results on these functions indicate that, compared with six state-of-the-art decomposition algorithms, SVG possesses broader applicability and competitive efficiency. Furthermore, it can significantly enhance the optimization performance of CC.

Deep Neural Networks (DNNs) have achieved great success in many applications. The architectures of DNNs play a crucial role in their performance, which is usually manually designed with rich expertise. However, such a design process is labour intensive because of the trial-and-error process, and also not easy to realize due to the rare expertise in practice. Neural Architecture Search (NAS) is a type of technology that can design the architectures automatically. Among different methods to realize NAS, Evolutionary Computation (EC) methods have recently gained much attention and success. Unfortunately, there has not yet been a comprehensive summary of the EC-based NAS algorithms. This paper reviews over 200 papers of most recent EC-based NAS methods in light of the core components, to systematically discuss their design principles as well as justifications on the design. Furthermore, current challenges and issues are also discussed to identify future research in this emerging field.

The theory of evolutionary computation for discrete search spaces has made significant of progress in the last ten years. This survey summarizes some of the most important recent results in this research area. It discusses fine-grained models of runtime analysis of evolutionary algorithms, highlights recent theoretical insights on parameter tuning and parameter control, and summarizes the latest advances for stochastic and dynamic problems. We regard how evolutionary algorithms optimize submodular functions and we give an overview over the large body of recent results on estimation of distribution algorithms. Finally, we present the state of the art of drift analysis, one of the most powerful analysis technique developed in this field.

A customized multi-objective evolutionary algorithm (MOEA) is proposed for the multi-objective flexible job shop scheduling problem (FJSP). It uses smart initialization approaches to enrich the first generated population, and proposes various crossover operators to create a better diversity of offspring. Especially, the MIP-EGO configurator, which can tune algorithm parameters, is adopted to automatically tune operator probabilities. Furthermore, different local search strategies are employed to explore the neighborhood for better solutions. In general, the algorithm enhancement strategy can be integrated with any standard EMO algorithm. In this paper, it has been combined with NSGA-III to solve benchmark multi-objective FJSPs, whereas an off-the-shelf implementation of NSGA-III is not capable of solving the FJSP. The experimental results show excellent performance with less computing budget.

In this paper a new population update rule for population based ant colony optimization (PACO) is proposed. PACO is a well known alternative to the standard ant colony optimization algorithm. The new update rule allows to weight different parts of the solutions. PACO with the new update rule is evaluated for the example of the single machine total weighted tardiness problem (SMTWTP). This is an NP -hard optimization problem where the aim is to schedule jobs on a single machine such that their total weighted tardiness is minimized. PACO with the new population update rule is evaluated with several benchmark instances from the OR-Library. Moreover, the impact of the weights of the jobs on the solutions in the population and on the convergence of the algorithm are analyzed experimentally. The results show that PACO with the new update rule has on average better solution quality than PACO with the standard update rule.

Recurrent neural networks (RNNs) provide state-of-the-art performances in a wide variety of tasks that require memory. These performances can often be achieved thanks to gated recurrent cells such as gated recurrent units (GRU) and long short-term memory (LSTM). Standard gated cells share a layer internal state to store information at the network level, and long term memory is shaped by network-wide recurrent connection weights. Biological neurons on the other hand are capable of holding information at the cellular level for an arbitrary long amount of time through a process called bistability. Through bistability, cells can stabilize to different stable states depending on their own past state and inputs, which permits the durable storing of past information in neuron state. In this work, we take inspiration from biological neuron bistability to embed RNNs with long-lasting memory at the cellular level. This leads to the introduction of a new bistable biologically-inspired recurrent cell that is shown to strongly improves RNN performance on time-series which require very long memory, despite using only cellular connections (all recurrent connections are from neurons to themselves, i.e. a neuron state is not influenced by the state of other neurons). Furthermore, equipping this cell with recurrent neuromodulation permits to link them to standard GRU cells, taking a step towards the biological plausibility of GRU.

The problem of catastrophic forgetting can be traced back to the 1980s, but it has not been completely solved. Since human brains are good at continual lifelong learning, brain-inspired methods may provide solutions to this problem. The end result of learning different objects in different categories is the formation of concepts in the brain. Experiments showed that concepts are likely encoded by concept cells in the medial temporal lobe (MTL) of the human brain. Furthermore, concept cells encode concepts sparsely and are responsive to multi-modal stimuli. However, it is unknown how concepts are formed in the MTL. Here we assume that the integration of audio and visual perceptual information in the MTL during learning is a crucial step to form concepts and make continual learning possible, and we propose a biological plausible audio-visual integration model (AVIM), which is a spiking neural network with multi-compartmental neuron model and a calcium based synaptic tagging and capture plasticity model, as a possible mechanism of concept formation. We then build such a model and run on different datasets to test its ability of continual learning. Our simulation results show that the AVIM not only achieves state-of-the-art performance compared with other advanced methods but also the output of AVIM for each concept has stable representations during the continual learning process. These results support our assumption that concept formation is essential for continuous lifelong learning, and suggest the AVIM we propose here is a possible mechanism of concept formation, and hence is a brain-like solution to the problem of catastrophic forgetting.