Artificial intelligence (AI) was a key topic at both MEDICA and the RSNA conference this year. But what are its applications in healthcare in general and radiology in particular? And what are the barriers? Dr. Michael Forsting, director of the Institute of Diagnostic and Interventional Radiology and Neuroradiology at Essen University Hospital in Germany talked to MedicalExpo e-magazine about his experiences with AI. MedicalExpo e-magazine: What are the major challenges facing AI in healthcare?

To fans of science fiction, artificial intelligence may remind them of robots like C-3PO, the loquacious but harmless golden droid in Star Wars, or Skynet in the Terminator movies, a calculating sentient computer that subjugated mankind. But AI is more than just a machine with human-level intelligence scientists hope they could one day create. It is a set of algorithms and technologies that is already powering many tasks in everyday life. Get our free ebook on how the Soviet Union became Putin's Russia. Chatbots that converse with you in Yahoo, Facebook and other sites use AI.

Computers might soon translate between many more languages. Automatic language translation has come a long way, thanks to neural networks--computer algorithms that take inspiration from the human brain. But training such networks requires an enormous amount of data: millions of sentence-by-sentence translations to demonstrate how a human would do it. Now, two new papers show that neural networks can learn to translate with no parallel texts--a surprising advance that could make documents in many languages more accessible. "Imagine that you give one person lots of Chinese books and lots of Arabic books--none of them overlapping--and the person has to learn to translate Chinese to Arabic. That seems impossible, right?" says the first author of one study, Mikel Artetxe, a computer scientist at the University of the Basque Country (UPV) in San Sebastiàn, Spain.

With movie theaters full of films about rogue robots taking over the world, many fear the impact artificial intelligence (AI) might have on human life. In light of this year's Queen's Lecture at the Technical University (TU) Berlin, DW has caught up with AI and engineering expert Zoubin Ghahramani. We asked him about human and artificial intelligence, machine learning and what he thinks our future with AI might look like. DW: Professor Ghahramani, before we speak about artificial intelligence and machine learning, could you define human intelligence for us? Zoubin Ghahramani: When they hear the word'intelligence' people often think about the differences between individual humans, but actually the more interesting question is'how are we different from other animals, plants and computers?'

We consider the problem of winner determination under Chamberlin--Courant's multiwinner voting rule with approval utilities. This problem is equivalent to the well-known NP-complete MaxCover problem and, so, the best polynomial-time approximation algorithm for it has approximation ratio 1 - 1/e. We show exponential-time/FPT approximation algorithms that, on one hand, achieve arbitrarily good approximation ratios and, on the other hand, have running times much better than known exact algorithms. We focus on the cases where the voters have to approve of at most/at least a given number of candidates.

AI developments on smart cities, if not critical, risk making a flawed urban model more efficient. Instead, we suggest that AI should challenge the mainstream techno-optimistic approach to solving urban problems by dialoguing with other academic fields, questioning the dominant urban paradigm, and creating transformative solutions. We claim that doing differently, rather than doing better, may be smarter for cities and the common good. This article is part of the special track on AI and Society.

We explore whether useful temporal neural generative models can be learned from sequential data without back-propagation through time. We investigate the viability of a more neurocognitively-grounded approach in the context of unsupervised generative modeling of sequences. Specifically, we build on the concept of predictive coding, which has gained influence in cognitive science, in a neural framework. To do so we develop a novel architecture, the Temporal Neural Coding Network, and its learning algorithm, Discrepancy Reduction. The underlying directed generative model is fully recurrent, meaning that it employs structural feedback connections and temporal feedback connections, yielding information propagation cycles that create local learning signals. This facilitates a unified bottom-up and top-down approach for information transfer inside the architecture. Our proposed algorithm shows promise on the bouncing balls generative modeling problem. Further experiments could be conducted to explore the strengths and weaknesses of our approach.

In this paper, we show that the recent integration of statistical models with deep recurrent neural networks provides a new way of formulating volatility (the degree of variation of time series) models that have been widely used in time series analysis and prediction in finance. The model comprises a pair of complementary stochastic recurrent neural networks: the generative network models the joint distribution of the stochastic volatility process; the inference network approximates the conditional distribution of the latent variables given the observables. Our focus here is on the formulation of temporal dynamics of volatility over time under a stochastic recurrent neural network framework. Experiments on real-world stock price datasets demonstrate that the proposed model generates a better volatility estimation and prediction that outperforms stronge baseline methods, including the deterministic models, such as GARCH and its variants, and the stochastic MCMC-based models, and the Gaussian-process-based, on the average negative log-likelihood measure.

In this paper we use information-theoretic measures to provide a theory and tools to analyze the flow of information from a discrete, multivariate source of information $\overline X$ to a discrete, multivariate sink of information $\overline Y$ joined by a distribution $P_{\overline X \overline Y}$. The first contribution is a decomposition of the maximal potential entropy of $(\overline X, \overline Y)$ that we call a balance equation, that can also be split into decompositions for the entropies of $\overline X$ and $\overline Y$ respectively. Such balance equations accept normalizations that allow them to be represented in de Finetti entropy diagrams, our second contribution. The most important of these, the aggregate Channel Multivariate Entropy Triangle CMET is an exploratory tool to assess the efficiency of multivariate channels. We also present a practical contribution in the application of these balance equations and diagrams to the assessment of information transfer efficiency for PCA and ICA as feature transformation and selection procedures in machine learning applications.

In statistical dialogue management, the dialogue manager learns a policy that maps a belief state to an action for the system to perform. Efficient exploration is key to successful policy optimisation. Current deep reinforcement learning methods are very promising but rely on epsilon-greedy exploration, thus subjecting the user to a random choice of action during learning. Alternative approaches such as Gaussian Process SARSA (GPSARSA) estimate uncertainties and are sample efficient, leading to better user experience, but on the expense of a greater computational complexity. This paper examines approaches to extract uncertainty estimates from deep Q-networks (DQN) in the context of dialogue management. We perform an extensive benchmark of deep Bayesian methods to extract uncertainty estimates, namely Bayes-By-Backprop, dropout, its concrete variation, bootstrapped ensemble and alpha-divergences, combining it with DQN algorithm.