Have you put a bet on the FIFA World Cup? If yes, the chances are you've made a pretty educated guess, right? You know which team has the strongest players or most favourable odds. Or maybe you've put some cash on your country's team, (which normally I'd avoid England, but given their recent performance, I could be wrong to!) Either way, you might be best casting your bets in line with San Francisco based Unanimous AI. They use a technology called Swarm AI - algorithms modelled on swarms in nature that amplifies human intelligence. By using human intelligence and artificial intelligence together, they can predict outcomes better than humans or AI acting alone.

The most data-efficient algorithms for reinforcement learning in robotics are model-based policy search algorithms, which alternate between learning a dynamical model of the robot and optimizing a policy to maximize the expected return given the model and its uncertainties. However, the current algorithms lack an effective exploration strategy to deal with sparse or misleading reward scenarios: if they do not experience any state with a positive reward during the initial random exploration, it is very unlikely to solve the problem. Here, we propose a novel model-based policy search algorithm, Multi-DEX, that leverages a learned dynamical model to efficiently explore the task space and solve tasks with sparse rewards in a few episodes. To achieve this, we frame the policy search problem as a multi-objective, model-based policy optimization problem with three objectives: (1) generate maximally novel state trajectories, (2) maximize the expected return and (3) keep the system in state-space regions for which the model is as accurate as possible. We then optimize these objectives using a Pareto-based multi-objective optimization algorithm. The experiments show that Multi-DEX is able to solve sparse reward scenarios (with a simulated robotic arm) in much lower interaction time than VIME, TRPO, GEP-PG, CMA-ES and Black-DROPS.

The effort devoted to hand-crafting image classifiers has motivated the use of architecture search to discover them automatically. Although evolutionary algorithms have been repeatedly applied to architecture search, the architectures thus discovered have remained inferior to human-crafted ones. Here we show for the first time that artificially-evolved architectures can match or surpass human-crafted and RL-designed image classifiers. In particular, our models---named AmoebaNets---achieved a state-of-the-art accuracy of 97.87% on CIFAR-10 and top-1 accuracy of 83.1% on ImageNet. Among mobile-size models, an AmoebaNet with only 5.1M parameters also achieved a state-of-the-art top-1 accuracy of 75.1% on ImageNet. We also compared this method against strong baselines. Finally, we performed platform-aware architecture search with evolution to find a model that trains quickly on Google Cloud TPUs. This method produced an AmoebaNet that won the Stanford DAWNBench competition for lowest ImageNet training cost.

Formalizing self reproduction in dynamical hierarchies is one of the important problems in Artificial Life (AL) studies. We study, in this paper, an inductively defined algebraic framework for self reproduction on macroscopic organizational levels under dynamical system setting for simulated AL models and explore some existential results. Starting with defining self reproduction for atomic entities we define self reproduction with possible mutations on higher organizational levels in terms of hierarchical sets and the corresponding inductively defined `meta' - reactions. We introduce constraints to distinguish a collection of entities from genuine cases of emergent organizational structures.

The scope of the Baldwin effect was recently called into question by two papers that closely examined the seminal work of Hinton and Nowlan. To this date there has been no demonstration of its necessity in empirically challenging tasks. Here we show that the Baldwin effect is capable of evolving few-shot supervised and reinforcement learning mechanisms, by shaping the hyperparameters and the initial parameters of deep learning algorithms. Furthermore it can genetically accommodate strong learning biases on the same set of problems as a recent machine learning algorithm called MAML "Model Agnostic Meta-Learning" which uses second-order gradients instead of evolution to learn a set of reference parameters (initial weights) that can allow rapid adaptation to tasks sampled from a distribution. Whilst in simple cases MAML is more data efficient than the Baldwin effect, the Baldwin effect is more general in that it does not require gradients to be backpropagated to the reference parameters or hyperparameters, and permits effectively any number of gradient updates in the inner loop. The Baldwin effect learns strong learning dependent biases, rather than purely genetically accommodating fixed behaviours in a learning independent manner.

Here on Earth, human settlements have thrived for so long due to two very important truths: we evolved on this planet and we survived by supporting each other. Our settlement history has not been perfect1. But as humans, our population vitality is a result of many individuals working to support the civilization in one large positive feedback loop of survival. The space-based human settlements of the future, however, will require advanced technology to continue this trend. Settlements on Mars, or even Earth's moon, Luna, will require human-machine teaming between the occupants of the settlement and artificial intelligence, or AI, to augment their skills and knowledge.

Genetic Algorithms (GAs) are used to solve search and optimization problems in which an optimal solution can be found using an iterative process with probabilistic and non-deterministic transitions. However, depending on the problem's nature, the time required to find a solution can be high in sequential machines due to the computational complexity of genetic algorithms. This work proposes a parallel implementation of a genetic algorithm on field-programmable gate array (FPGA). Optimization of the system's processing time is the main goal of this project. Results associated with the processing time and area occupancy (on FPGA) for various population sizes are analyzed. Studies concerning the accuracy of the GA response for the optimization of two variables functions were also evaluated for the hardware implementation. However, the high-performance implementation proposes in this paper is able to work with more variable from some adjustments on hardware architecture.

This paper addresses batch scheduling at a back-end semiconductor plant of Nexperia. This complex manufacturing environment is characterized by a large product and batch size variety, numerous parallel machines with large capacity differences, sequence and machine dependent setup times and machine eligibility constraints. A hybrid genetic algorithm is proposed to improve the scheduling process, the main features of which are a local search enhanced crossover mechanism, two additional fast local search procedures and a user-controlled multi-objective fitness function. Testing with real-life production data shows that this multi-objective approach can strike the desired balance between production time, setup time and tardiness, yielding high-quality practically feasible production schedules.

Incremental learning from non-stationary data poses special challenges to the field of machine learning. Although new algorithms have been developed for this, assessment of results and comparison of behaviors are still open problems, mainly because evaluation metrics, adapted from more traditional tasks, can be ineffective in this context. Overall, there is a lack of common testing practices. This paper thus presents a testbed for incremental non-stationary learning algorithms, based on specially designed synthetic datasets. Also, test results are reported for some well-known algorithms to show that the proposed methodology is effective at characterizing their strengths and weaknesses. It is expected that this methodology will provide a common basis for evaluating future contributions in the field.

Apart from few exceptions, the mathematical runtime analysis of evolutionary algorithms is mostly concerned with expected runtimes. In this work, we argue that stochastic domination is a notion that should be used more frequently in this area. Stochastic domination allows to formulate much more informative performance guarantees, it allows to decouple the algorithm analysis into the true algorithmic part of detecting a domination statement and the probability-theoretical part of deriving the desired probabilistic guarantees from this statement, and it helps finding simpler and more natural proofs. As particular results, we prove a fitness level theorem which shows that the runtime is dominated by a sum of independent geometric random variables, we prove the first tail bounds for several classic runtime problems, and we give a short and natural proof for Witt's result that the runtime of any $(\mu,p)$ mutation-based algorithm on any function with unique optimum is subdominated by the runtime of a variant of the \oea on the \onemax function. As side-products, we determine the fastest unbiased (1+1) algorithm for the \leadingones benchmark problem, both in the general case and when restricted to static mutation operators, and we prove a Chernoff-type tail bound for sums of independent coupon collector distributions.