Learning classifier systems are adaptive learning systems which have been widely applied in a multitude of application domains. However, there are still some generalization problems unsolved. The hurdle is that fitness and niching pressures are difficult to balance. Here, a new algorithm called Self Organizing Classifiers is proposed which faces this problem from a different perspective. Instead of balancing the pressures, both pressures are separated and no balance is necessary. In fact, the proposed algorithm possesses a dynamical population structure that self-organizes itself to better project the input space into a map. The niched fitness concept is defined along with its dynamical population structure, both are indispensable for the understanding of the proposed method. Promising results are shown on two continuous multi-step problems. One of which is yet more challenging than previous problems of this class in the literature.

Noname manuscript No. (will be inserted by the editor) Abstract Learning classifier systems are evolutionary machine learning algorithms, flexible enough to be applied toreinforcement, supervised and unsupervised learning problems with good performance. Recently, self organizing classifierswere proposed which are similar to learning classifier systems but have the advantage that in its structured population no balance between niching and fitness pressure is necessary. However, more tests and analysis are required to verify its benefits. Here, a variation of the first algorithm is proposed which uses a parameterless self organizing map (SOM). This algorithm isapplied in challenging problems such as big, noisy as well as dynamically changing continuous inputaction mazes(growing and compressing mazes are included) withgood performance. Moreover, a genetic operator is proposed which utilizes the topological information ofthe SOM's population structure, improving the results. Thus, the first steps in structured evolutionary machinelearning are shown, nonetheless, the problems faced are more difficult than the state-of-art continuous input-action multi-step ones. 1 Introduction Learning Classifier Systems (LCS) are several algorithms inspired by evolution [29],[20]. Different from most reinforcement learning algorithms, however, LCS algorithms do not use state-action lookup tables to predict payoff. In this manner, the difficulties that arrive from complex problems, wherea large number of states and/or actions are required, can be avoided. Oneway of solving this problem is to separate a fitness defined on a niche from fitnesses defined on other niches (i.e., having a good fitness on other niches would not influence the present niche).

SoftBank Group Corp. is introducing a new robot that, unlike the talkative Pepper, skips the chit chat and just mops the floor. Whiz, an autonomous floor-cleaning machine for businesses, will go on sale in Japan in February, the company announced Monday. The 32-kg machine is powered by self-driving software and an array of sensors from Brain Corp., a San Diego-based startup that is part of SoftBank's $100 billion Vision Fund. It will be available for rent for ¥25,000 a month. Pepper, SoftBank's first foray into robotics, was marketed as a companion in the home and as a sales assistant on the shop floor.

TEL AVIV, November 15, 2018 – Seebo today announced the launch of its unique process-based artificial intelligence (AI) technology. The new AI-based capabilities for production line data introduce unmatched accuracy and ease of use of the company's predictive quality, predictive maintenance and production line intelligence solutions. Process manufacturers today face rising demands on production capacity and continuous disruptions that affect uptime, quality, and throughput. Increasingly, they are turning to machine-generated data to investigate and solve their production line problems. But finding meaningful insights entails applying sophisticated machine learning technologies to a carefully engineered big data repository – a process beyond the technical and financial reach of most manufacturers.

Japanese billionaire Masayoshi Son, the founder and chief executive of technology conglomerate SoftBank Group Corp, is known for making solid bets in China's hi-tech sector. Around 18 years ago, Son's company invested US$20 million in a small Chinese online retail platform that rapidly grew to become e-commerce giant Alibaba Group Holding. Son in July invited the heads of fast-rising Chinese companies Ping An Good Doctor and Didi Chuxing to a party he hosted in Tokyo, in a testament to how far these two firms have grown since SoftBank invested in them. Wang Tao, the founder, chairman and chief executive of Ping An Good Doctor, acknowledged Son's contribution amid the Hong Kong-listed online health care provider's efforts to innovate and extend its operations outside the mainland. "Mr Son helped us a lot in our international expansion," said Wang in an interview with the South China Morning Post on the sidelines of the fifth World Internet Conference held earlier this month in Wuzhen, a town in China's eastern coastal province of Zhejiang.

Mark Zuckerberg fiercely defended Facebook in a question-and-answer session with employees on Friday afternoon, pushing back against criticism of the company in the wake of a New York Times investigation into how it reacted to Russian influence operations. In an hour-long video-conference broadcast to Facebook offices around the world, Mr Zuckerberg responded to questions from employees on a range of topics, from Facebook's behaviour over the past 18 months to how it should handle leaks to the media, according to three people familiar with the discussion but not willing to discuss it publicly because it was a private meeting. The idea that Facebook tried to "cover up anything" was wrong, an impassioned Mr Zuckerberg said, using an expletive in his response, according to these people. Some employees responded with muted applause and cheers. The session came at a fraught time for the social network, as executives mobilised to deal with a torrent of criticism of the company.

Representation learning is a central challenge across a range of machine learning areas. In reinforcement learning, effective and functional representations have the potential to tremendously accelerate learning progress and solve more challenging problems. Most prior work on representation learning has focused on generative approaches, learning representations that capture all underlying factors of variation in the observation space in a more disentangled or well-ordered manner. In this paper, we instead aim to learn functionally salient representations: representations that are not necessarily complete in terms of capturing all factors of variation in the observation space, but rather aim to capture those factors of variation that are important for decision making -- that are "actionable." These representations are aware of the dynamics of the environment, and capture only the elements of the observation that are necessary for decision making rather than all factors of variation, without explicit reconstruction of the observation. We show how these representations can be useful to improve exploration for sparse reward problems, to enable long horizon hierarchical reinforcement learning, and as a state representation for learning policies for downstream tasks. We evaluate our method on a number of simulated environments, and compare it to prior methods for representation learning, exploration, and hierarchical reinforcement learning.

Safe, controllable and credible artificial intelligence has been the goal which the humanity has been pursuing. In the field of deep learning, in order to achieve this goal, it is needed for learning algorithm to really interact with the humanity and it is also indispensable for the learning algorithm to have the ability to correct errors, so as to avoid a prediction model with serious errors caused by unnecessary deviation in training data. So, it is necessary to establish a learning algorithm for capturing and learning causal relationships in the world around us. However, recently, all of this is out of reach. The reason is that the prediction model and its training process are not yet understood by human being. In other words, there is a gap between the deep learning model and the cognitive modes from human being. For shrinking the gap, two general interpretation methods about deep learning model were identified by Lipton [1]: posting interpretation and transparent interpretation. For deep learning, the current mainstream methods are mainly from three aspects: hidden layer analysis method [2-4], simulation model method [5], attention mechanism[6-8]. We posits that how to make the prediction model and training process understood by us ascribe to an optimization problem which can promote the interpretability of the prediction model and make the model more suitable to its causality or discover faults in the causality.

Representation learning of pedestrian trajectories transforms variable-length timestamp-coordinate tuples of a trajectory into a fixed-length vector representation that summarizes spatiotemporal characteristics. It is a crucial technique to connect feature-based data mining with trajectory data. Trajectory representation is a challenging problem, because both environmental constraints (e.g., wall partitions) and temporal user dynamics should be meticulously considered and accounted for. Furthermore, traditional sequence-to-sequence autoencoders using maximum log-likelihood often require dataset covering all the possible spatiotemporal characteristics to perform well. This is infeasible or impractical in reality. We propose TREP, a practical pedestrian trajectory representation learning algorithm which captures the environmental constraints and the pedestrian dynamics without the need of any training dataset. By formulating a sequence-to-sequence autoencoder with a spatial-aware objective function under the paradigm of actor-critic reinforcement learning, TREP intelligently encodes spatiotemporal characteristics of trajectories with the capability of handling diverse trajectory patterns. Extensive experiments on both synthetic and real datasets validate the high fidelity of TREP to represent trajectories.

In some developing countries, slum residents make up for more than half of the population and lack reliable sanitation services, clean water, electricity, other basic services. Thus, slum rehabilitation and improvement is an important global challenge, and a significant amount of effort and resources have been put into this endeavor. These initiatives rely heavily on slum mapping and monitoring, and it is essential to have robust and efficient methods for mapping and monitoring existing slum settlements. In this work, we introduce an approach to segment and map individual slums from satellite imagery, leveraging regional convolutional neural networks for instance segmentation using transfer learning. In addition, we also introduce a method to perform change detection and monitor slum change over time. We show that our approach effectively learns slum shape and appearance, and demonstrates strong quantitative results, resulting in a maximum AP of 80.0.