Self-driving cars are being developed by several major technology companies and carmakers. When a driver slams on the brakes to avoid hitting a pedestrian crossing the road illegally, she is making a moral decision that shifts risk from the pedestrian to the people in the car. Self-driving cars might soon have to make such ethical judgments on their own -- but settling on a universal moral code for the vehicles could be a thorny task, suggests a survey of 2.3 million people from around the world. The largest ever survey of machine ethics1, published today in Nature, finds that many of the moral principles that guide a driver's decisions vary by country. For example, in a scenario in which some combination of pedestrians and passengers will die in a collision, people from relatively prosperous countries with strong institutions were less likely to spare a pedestrian who stepped into traffic illegally.

Artificial intelligence has been a hot subject in real estate for a few years now and while some remain uncomfortable with the fast-growing trend, others are already saving time and money by using such technologies in their businesses. While AI has its challenges, the potential gains for the industry far outweigh them. In one of the opening panels of 2018's Expo Real, the yearly international real estate and investment conference taking place in München, Bastian Schulz, head of sales at Leverton; Sascha Donner, co-founder & head of product at startup Evana; Matthew Webster, CFO of Cloudscraper; and Nicolai Wendland, COO of 21st Real Estate, shared their vision on how AI could shape the industry. The highest goal for real estate professionals using AI seems to be the simplification of the acquisition process, at a larger scale. "There is no bigger vision than imagining the process of purchasing one property with one click, like we do with shares. We will be there soon, in a few years," Wendland said.

Choose "I don't think so." or "Yes, I think so.". Our world of technophiles is sometimes pushed to the extreme: many fans will defend the brands they adore to a point that even worries some sociologists. This is the power of marketing in its raw state, and it's neither new (as this 1994 article attests) nor truly surprising since current generations were born in it and are bathing in it without even realizing it. This is a worrying subject and techno-skeptics tend to rush into the same arguments: "technology companies are interested in nothing but themselves", "they don't want to help users as they claim, they just alienate them to keep them in their midst", etc. These points, which are generally well founded, still need to be qualified.

Researchers are planning to recreate the conditions of the lunar surface right here at home. A new facility in the works at ESA's Astronaut Centre in Cologne, Germany will soon serve as a three-part moon analogue environment on Earth, the agency announced this month. There, scientists will simulate lunar soil and a moon habitat, powered by systems that could one day be used to support a real base on the moon. Researchers are planning to recreate the conditions of the lunar surface right here at home. A new facility in the works at ESA's Astronaut Centre in Cologne, Germany will soon serve as a three-part moon analogue environment on Earth.

Generative adversarial networks (GANs) -- two-part neural networks consisting of generators that produce samples and discriminators that attempt to distinguish between the generated samples and real-world samples -- have been used to discover new drugs, create convincing photos of burgers and butterflies, and generate synthetic scans of brain cancer. And as a new paper published by Maastricht University in the Netherlands reveals, they're not half bad at generating logos, either. In research published on the preprint server Arxiv.org "Designing a logo is a long, complicated, and expensive process for any designer. However, recent advancements in generative algorithms provide models that could offer a possible solution," they wrote.

The Expectation-Maximization (EM) algorithm for mixture models often results in slow or invalid convergence. The popular convergence proof affirms that the likelihood increases with Q; Q is increasing in the M -step and non-decreasing in the E-step. The author found that (1) Q may and should decrease in some E-steps; (2) The Shannon channel from the E-step is improper and hence the expectation is improper. The author proposed the CM-EM algorithm (CM means Channel's Matching), which adds a step to optimize the mixture ratios for the proper Shannon channel and maximizes G, average log-normalized-likelihood, in the M-step. Neal and Hinton's Maximization-Maximization (MM) algorithm use F instead of Q to speed the convergence. Maximizing G is similar to maximizing F. The new convergence proof is similar to Beal's proof with the variational method. It first proves that the minimum relative entropy equals the minimum R-G (R is mutual information), then uses variational and iterative methods that Shannon et al. use for rate-distortion functions to prove the global convergence. Some examples show that Q and F should and may decrease in some E-steps. For the same example, the EM, MM, and CM-EM algorithms need about 36, 18, and 9 iterations respectively.

As opposed to standard empirical risk minimization (ERM), distributionally robust optimization aims to minimize the worst-case risk over a larger ambiguity set containing the original empirical distribution of the training data. In this work, we describe a minimax framework for statistical learning with ambiguity sets given by balls in Wasserstein space. In particular, we prove generalization bounds that involve the covering number properties of the original ERM problem. As an illustrative example, we provide generalization guarantees for transport-based domain adaptation problems where the Wasserstein distance between the source and target domain distributions can be reliably estimated from unlabeled samples.

We review the current state of automatic differentiation (AD) for array programming in machine learning (ML), including the different approaches such as operator overloading (OO) and source transformation (ST) used for AD, graph-based intermediate representations for programs, and source languages. Based on these insights, we introduce a new graph-based intermediate representation (IR) which specifically aims to efficiently support fully-general AD for array programming. Unlike existing dataflow programming representations in ML frameworks, our IR naturally supports function calls, higher-order functions and recursion, making ML models easier to implement. The ability to represent closures allows us to perform AD using ST without a tape, making the resulting derivative (adjoint) program amenable to ahead-of-time optimization using tools from functional language compilers, and enabling higher-order derivatives. Lastly, we introduce a proof of concept compiler toolchain called Myia which uses a subset of Python as a front end.

Recent progress on deep learning has made it possible to automatically transform the screenshot of Graphic User Interface (GUI) into code by using the encoder-decoder framework. While the commonly adopted image encoder (e.g., CNN network), might be capable of extracting image features to the desired level, interpreting these abstract image features into hundreds of tokens of code puts a particular challenge on the decoding power of the RNN-based code generator. Considering the code used for describing GUI is usually hierarchically structured, we propose a new attention-based hierarchical code generation model, which can describe GUI images in a finer level of details, while also being able to generate hierarchically structured code in consistency with the hierarchical layout of the graphic elements in the GUI. Our model follows the encoder-decoder framework, all the components of which can be trained jointly in an end-to-end manner. The experimental results show that our method outperforms other current state-of-the-art methods on both a publicly available GUI-code dataset as well as a dataset established by our own.

We address the problem of speech enhancement generalisation to unseen environments by performing two manipulations. First, we embed an additional recording from the environment alone, and use this embedding to alter activations in the main enhancement subnetwork. Second, we scale the number of noise environments present at training time to 16,784 different environments. Experiment results show that both manipulations reduce word error rates of a pretrained speech recognition system and improve enhancement quality according to a number of performance measures. Specifically, our best model reduces the word error rate from 34.04% on noisy speech to 15.46% on the enhanced speech. Enhanced audio samples can be found in https://speechenhancement.page.link/samples.