In his famous Foundation novels, written in the 1940s, Isaac Asimov imagined something called psychohistory, a discipline which used statistical modelling and a detailed understanding of the mind to predict the future. To Asimov, this idea seemed so futuristic that he placed it 20,000 years in the future. Data scientists are using the latest AI to model and predict the behaviour of crowds and other large groups of people in ways that help authorities plan the provision of services. In Japan, for example, the Kanagawa Prefectural Police is using AI to analyse variables such as weather data, crowd dynamics and even social media activity to predict crime patterns and deploy officers. This is just one of the ways in which AI is making us smarter.

Editor's Note: We know artificial intelligence will change the very nature of war--but we don't know how. The United States, China, and other powers recognize this transformative potential and, even as they seek to exploit it, fear that others will gain the upper hand in an artificial intelligence arms race. My Brookings colleague Chris Meserole describes how artificial intelligence might produce a new security dilemma and proposes several ways to mitigate the risk. Recent breakthroughs in machine learning and artificial intelligence (A.I.) have prompted breathless speculation about their national security applications. Yet most of that work has focused narrowly on their implications for autonomous weapons systems, rather than on the broader security environment.

Scientists have developed an artificial intelligence system that can help better understand what triggers religious violence. The study, published in The Journal for Artificial Societies and Social Stimulation, focuses on two cases of extreme violence, firstly, the conflict commonly referred to as the Northern Ireland Troubles, which is regarded as one of the most violent periods in Irish history. The conflict, involving the British army and various Republican and Loyalist paramilitary groups, spanned three decades, claimed the lives of approximately 3,500 people and saw a further 47,000 injured. Although a much shorter period of tension, the 2002 Gujarat riots of India were equally devastating. The three-day period of inter-communal violence between the Hindu and Muslim communities in the western Indian state of Gujarat, began when a Sabarmarti Express train filled with Hindu pilgrims, stopped in the, predominantly Muslim town of Godhra, and ended with the deaths of more than 2,000 people.

Indian Institute of Technology Madras undertook a major effort to give a boost to the Artificial Intelligence (AI) and Machine Learning (ML) sectors in Chennai. The Robert Bosch Center for Data Science and Artificial Intelligence, IIT Madras, organized the'Artificial Intelligence and Machine Learning Conclave' focused on understand cutting-edge technology and innovation in the field with participation from top technology firms and think-tanks including Google, Amazon, Foxconn and TVS group among others. Prof Bhaskar Ramamurthi, Director, IIT Madras, inaugurated the Conclave, which was held on 23rd October 2018. The conclave aimed at generating a greater realization of the AI/ML ecosystem in and around Chennai and facilitated the stakeholders to have a brainstorming session about the needs for this ecosystem to thrive and grow further. This event for the first time brought together a significant number of AI/ML deep technology start-ups in Chennai in a single platform.

While deep neural networks have achieved groundbreaking prediction results in many tasks, there is a class of data where existing architectures are not optimal -- sequences of probability distributions. Performing forward prediction on sequences of distributions has many important applications. However, there are two main challenges in designing a network model for this task. First, neural networks are unable to encode distributions compactly as each node encodes just a real value. A recent work of Distribution Regression Network (DRN) solved this problem with a novel network that encodes an entire distribution in a single node, resulting in improved accuracies while using much fewer parameters than neural networks. However, despite its compact distribution representation, DRN does not address the second challenge, which is the need to model time dependencies in a sequence of distributions. In this paper, we propose our Recurrent Distribution Regression Network (RDRN) which adopts a recurrent architecture for DRN. The combination of compact distribution representation and shared weights architecture across time steps makes RDRN suitable for modeling the time dependencies in a distribution sequence. Compared to neural networks and DRN, RDRN achieves the best prediction performance while keeping the network compact.

We present and study models of adversarial online learning where the feedback observed by the learner is noisy, and the feedback is either full information feedback or bandit feedback. Specifically, we consider binary losses xored with the noise, which is a Bernoulli random variable. We consider both a constant noise rate and a variable noise rate. Our main results are tight regret bounds for learning with noise in the adversarial online learning model.

This study introduces a stochastic predictive-coding RNN model that can learn to extract probabilistic structures hidden in fluctuating temporal patterns by dynamically changing the uncertainty of latent variables. The learning process of the model involves maximizing the lower bound on the marginal likelihood of the sequential data, which consists of two terms. The first one is the expectation of prediction errors and the second one is the divergence of the prior and the approximated posterior. The main focus in the current study is to examine how weighting of the second term during learning affects the way of internally representing the uncertainty hidden in the sequence data. The simulation experiment on learning a simple probabilistic finite state machine demonstrates that the estimation of uncertainty in the latent variable approaches zero at each time step and that the network imitates the probabilistic structure of the target sequences by developing deterministic chaos in the case of the high weighting. On the contrary, in the case of the low weighting, the estimate of uncertainty increases significantly because of developing a random process in the network. The analysis shows that generalization in learning is most successful between these two extremes. Qualitatively, the same property has been observed in a trail of learning more complex sequence data consisting of probabilistic transitions between a set of hand-drawn primitive patterns using the model extended with hierarchy.

In this article we present a function fitting method, which is a convex minimization problem and can be solved using a gradient descent algorithm. We also provide some analysis on the fitness of the function to the data. The function fitting problem is also shown to be a solution of a linear, weak pde which contains some global terms. We describe a simple numerical solution using a gradient descent algorithm, that converges uniformly to the actual solution.As the minimization problem is also that of a quadratic form, there also exists a numerical method using linear algebra.

Emily Parker, who covered China for the Wall Street Journal, is the author of "Now I Know Who My Comrades Are: Voices From the Internet Underground." Silicon Valley was once able to write off Chinese tech companies as mere copycats. The big American players, from Twitter to Facebook to Google, all had a Chinese impersonator. But the rise of hugely successful Chinese messaging apps like WeChat -- not to mention all the U.S. tech companies that failed in China -- now make clear that the nation's tech companies should not be underestimated. In his book "AI Superpowers," Kai-Fu Lee, a well-known artificial-intelligence expert, venture capitalist and former president of Google China, argues that China and Silicon Valley will lead the world in AI.

The Friday event was hosted by the Cabinet Office and others. The hand-shaped robot, developed by Tohoku University, has fingers consisting of small ball-like parts, operated through wires running through its length. The robot, which features enhanced fire resistance, is expected to be useful in the event of a plant fire, according to the university. At the test event Friday, the robot removed gas cylinders and rubble from a fire. The cybersuit, developed by the university and others, is equipped with a camera and a GPS device.