Natural learners must compute an estimate of future outcomes that follow from a stimulus in continuous time. Critically, the learner cannot in general know a priori the relevant time scale over which meaningful relationships will be observed. Widely used reinforcement learning algorithms discretize continuous time and use the Bellman equation to estimate exponentially-discounted future reward. However, exponential discounting introduces a time scale to the computation of value. Scaling is a serious problem in continuous time: efficient learning with scaled algorithms requires prior knowledge of the relevant scale. That is, with scaled algorithms one must know at least part of the solution to a problem prior to attempting a solution. We present a computational mechanism, developed based on work in psychology and neuroscience, for computing a scale-invariant timeline of future events. This mechanism efficiently computes a model for future time on a logarithmically-compressed scale, and can be used to generate a scale-invariant power-law-discounted estimate of expected future reward. Moreover, the representation of future time retains information about what will happen when, enabling flexible decision making based on future events. The entire timeline can be constructed in a single parallel operation.

Toribash is a MuJoCo-like environment of two humanoid character fighting each other hand-to-hand, controlled by changing states of body joints. Competitive nature of Toribash lends itself to two-agent experiments, and active player-base can be used for human baselines. This white paper describes the environment with its pros, cons and limitations as well experimentally show ToriLLE's applicability as a learning environment by successfully training reinforcement learning agents that improved over time. The code is available at https: //github.com/Miffyli/ToriLLE.

Although there is an emerging trend towards generating embeddings for primarily unstructured data, and recently for structured data, there is not yet any systematic suite for measuring the quality of embeddings. This deficiency is further sensed with respect to embeddings generated for structured data because there are no concrete evaluation metrics measuring the quality of encoded structure as well as semantic patterns in the embedding space. In this paper, we introduce a framework containing three distinct tasks concerned with the individual aspects of ontological concepts: (i) the categorization aspect, (ii) the hierarchical aspect, and (iii) the relational aspect. Then, in the scope of each task, a number of intrinsic metrics are proposed for evaluating the quality of the embeddings. Furthermore, w.r.t. this framework multiple experimental studies were run to compare the quality of the available embedding models. Employing this framework in future research can reduce misjudgment and provide greater insight about quality comparisons of embeddings for ontological concepts.

Generating large quantities of quality labeled data in medical imaging is very time consuming and expensive. The performance of supervised algorithms for various tasks on imaging has improved drastically over the years, however the availability of data to train these algorithms have become one of the main bottlenecks for implementation. To address this, we propose a semi-supervised learning method where pseudo-negative labels from unlabeled data are used to further refine the performance of a pulmonary nodule detection network in chest radiographs. After training with the proposed network, the false positive rate was reduced to 0.1266 from 0.4864 while maintaining sensitivity at 0.89.

Multi-Task Learning (MTL) is a learning paradigm in machine learning and its aim is to leverage useful information contained in multiple related tasks to help improve the generalization performance of all the tasks. In this paper, we give a survey for MTL. First, we classify different MTL algorithms into several categories, including feature learning approach, low-rank approach, task clustering approach, task relation learning approach, and decomposition approach, and then discuss the characteristics of each approach. In order to improve the performance of learning tasks further, MTL can be combined with other learning paradigms including semi-supervised learning, active learning, unsupervised learning, reinforcement learning, multi-view learning and graphical models. When the number of tasks is large or the data dimensionality is high, batch MTL models are difficult to handle this situation and online, parallel and distributed MTL models as well as dimensionality reduction and feature hashing are reviewed to reveal their computational and storage advantages. Many real-world applications use MTL to boost their performance and we review representative works. Finally, we present theoretical analyses and discuss several future directions for MTL.

Amid Beijing's push to turn the country into an artificial intelligence (AI) powerhouse and embed the technology in all facets of life, some Chinese entrepreneurs are taking the expertise to a new frontier: sex dolls. WMDOLL, one of China's biggest sex doll makers, which is based in the southeastern province of Guangdong, launched what it calls AI-powered dolls at end of 2016 that offer features ranging from simple conversation to moving eyes, arms and torsos. Customers can personalize their dolls by choosing various appearance options including height, hairstyle and eye color. AI features on the dolls are still very basic: they can answer questions but cannot hold longer conversations. The doll uses vocabulary by connecting to a database supported by Chinese tech giant Baidu.

Reprinted with permission from Quanta Magazine's Abstractions blog. Occasionally the machine may rattle reality enough to have a few of those collisions generate something that's never been seen before. But because these events are by their nature a surprise, physicists don't know exactly what to look for. They worry that in the process of winnowing their data from those billions of collisions to a more manageable number, they may be inadvertently deleting evidence for new physics. "We're always afraid we're throwing the baby away with the bathwater," said Kyle Cranmer, a particle physicist at New York University who works with the ATLAS experiment at CERN.

Since machine learning (ML) models encompass a large amount of data besides an intensive analysis in its algorithms, it is ideal to bring up an optimal solution environment in its efficacy. This is where dynamic programming comes into the picture. It is specifically used in the context of reinforcement learning (RL) applications in ML. It is also suitable for applications where decision processes are critical in a highly uncertain environment. In this article, we explore the nuances of dynamic programming with respect to ML. Dynamic Programming (DP) is one of the techniques available to solve self-learning problems.

Drones have quickly become a technology almost everyone is familiar with. Whether it's professional photographers capturing wild landscapes or amateurs racing mini quadcopters around their living room: drones are here to stay. But not all drones are the same. If your mission is to use the power of a drone to capture your most unmissable moments, then the Airlango Mystic Drone is your answer. The Airlango Mystic Drone sets itself apart from its competition by incorporating a range of technologies to make it easy for anyone to be a professional drone photography pilot.

In the trending era of artificial intelligence, chatbot market is witnessing extraordinary growth owing to the prevalence of messaging platforms, virtual assistants, and the efforts of various businesses to deliver prompt customer service. Built using AI and machine learning technologies, today's chatbots are characterized by the ability to respond differently to varied keywords, learn, and adapt their own responses to suit diverse situations. These features have proven to be ideal in leveraging the bots for real-time communications, handling customer enquiries, and other aspects of businesses to optimize overall customer satisfaction, pushing the global chatbot market revenue over USD 250 million in 2017. An undeniable fact regarding AI technology is the prolific speed of its evolution, which in the process is transforming the chatbot industry. From the basic rule-based chatbots programmed to handle uncomplicated customer queries to the cloud-based, advanced self-learning ones that can understand intent and modify their output, chatbots have been extensively adopted in segments such as ecommerce with continuous integration of business data and internal systems.