I have been invited to write a book chapter on lexical choice for translators (contact me if you want to see a preprint). To get acquainted on this audience different from my usual computer science I read a few papers on professional translators use of technology. Two of them are quite interesting and I recommend them not only because they make for a good read and they have implications outside translation: Translation Skill-sets in a Machine-translation Age by Anthony Pym (2013) and Is Machine Translation Post-editing Worth the Effort?: A Survey of Research into Post-editing and Effort by Maarit Koponen (2016). This search finished by reading a short ebook by researchers at the MIT Center for Digital Business titled Race Against the Machine: How the Digital Revolution Is Accelerating Innovation, Driving Productivity, and Irreversibly Transforming Employment and the Economy. In that book plus the papers there's this call for humans, if we want to remain employed, to hybridize our work and to seek out ways to work with the computer as some sort of partnership.

A study has found that people are inclined to trust chatbots with sensitive information and that they are open to receiving advice from these AI services. The "Humanity in the Machine" report --published by media agency Mindshare UK on Thursday -- urges brands to engage with customers through chatbots, which can be defined as artificial intelligence programmes that conduct conversations with humans through chat interfaces. In conjunction with Goldsmiths University and experts at the likes of IBM, Mindshare UK observed how consumers use chatbots when making a banking transaction. It also surveyed 1,000 UK smartphone users aged 18-65 on their attitudes towards chatbots. The study found that 63% of people would consider messaging an online chatbot to get in touch with a business or brand.

The U.S. announced Thursday the first fatality of a wreck involving a car in self-driving mode, the 40-year-old owner of a technology company who nicknamed his vehicle "Tessy" and had praised its sophisticated "Autopilot" system just one month earlier for preventing a collision on an interstate. The government said it is investigating the design and performance of the system aboard the Tesla Model S sedan. Joshua D. Brown, 40, of Canton, Ohio, died in the accident May 7 in Williston, Florida, when his car's cameras failed to distinguish the white side of a turning tractor-trailer from a brightly lit sky and didn't automatically activate its brakes, according to government records obtained Thursday. Frank Baressi, 62, the driver of the truck and owner of Okemah Express LLC, said the Tesla driver was "playing Harry Potter on the TV screen" at the time of the crash and driving so quickly that "he went so fast through my trailer I didn't see him." "It was still playing when he died and snapped a telephone pole a quarter mile down the road," Baressi told The Associated Press in an interview from his home in Palm Harbor, Florida.

FILE - In this Monday, April 25, 2016, file photo, a man sits behind the steering wheel of a Tesla Model S electric car on display at the Beijing International Automotive Exhibition in Beijing. Federal officials say the driver of a Tesla S sports car using the vehicle's "autopilot" automated driving system has been killed in a collision with a truck, the first U.S. self-driving car fatality. The National Highway Traffic Safety Administration said preliminary reports indicate the crash occurred when a tractor-trailer made a left turn in front of the Tesla at a highway intersection. NHTSA said the Tesla driver died due to injuries sustained in the crash, which took place on May 7 in Williston, Fla. (AP Photo/Mark Schiefelbein, File) FILE - In this Monday, April 25, 2016, file photo, a man sits behind the steering wheel of a Tesla Model S electric car on display at the Beijing International Automotive Exhibition in Beijing. Federal officials say the driver of a Tesla S sports car using the vehicle's "autopilot" automated driving system has been killed in a collision with a truck, the first U.S. self-driving car fatality.

What was the evolution of your interest in machine learning, and how did you zero-in on your Ph.D. work? Ilya Sutskever: I started my Ph.D. just before deep learning became a thing. I was working on a number of different projects, mostly centered around neural networks.

The purpose of the report is to detail the process of training an Artificial Neural Network to control a robot. This report will be divided into several sections. The goal of this report is to demonstrate the ability of an ANN to control a robot to stand. In the previous reports the GA was used to evolve an ideal Artificial Neural Network topology, which was then refined via backpropagation learning. For this report the same techniques will be applied to the process of training an ANN to control a simulated robot, referred to simBot in this report.

Early diagnosis, playing an important role in preventing progress and treating the Alzheimer\{'}s disease (AD), is based on classification of features extracted from brain images. The features have to accurately capture main AD-related variations of anatomical brain structures, such as, e.g., ventricles size, hippocampus shape, cortical thickness, and brain volume. This paper proposed to predict the AD with a deep 3D convolutional neural network (3D-CNN), which can learn generic features capturing AD biomarkers and adapt to different domain datasets. The 3D-CNN is built upon a 3D convolutional autoencoder, which is pre-trained to capture anatomical shape variations in structural brain MRI scans. Fully connected upper layers of the 3D-CNN are then fine-tuned for each task-specific AD classification. Experiments on the CADDementia MRI dataset with no skull-stripping preprocessing have shown our 3D-CNN outperforms several conventional classifiers by accuracy. Abilities of the 3D-CNN to generalize the features learnt and adapt to other domains have been validated on the ADNI dataset.

Brain networks in fMRI are typically identified using spatial independent component analysis (ICA), yet mathematical constraints such as sparse coding and positivity both provide alternate biologically-plausible frameworks for generating brain networks. Non-negative Matrix Factorization (NMF) would suppress negative BOLD signal by enforcing positivity. Spatial sparse coding algorithms ($L1$ Regularized Learning and K-SVD) would impose local specialization and a discouragement of multitasking, where the total observed activity in a single voxel originates from a restricted number of possible brain networks. The assumptions of independence, positivity, and sparsity to encode task-related brain networks are compared; the resulting brain networks for different constraints are used as basis functions to encode the observed functional activity at a given time point. These encodings are decoded using machine learning to compare both the algorithms and their assumptions, using the time series weights to predict whether a subject is viewing a video, listening to an audio cue, or at rest, in 304 fMRI scans from 51 subjects. For classifying cognitive activity, the sparse coding algorithm of $L1$ Regularized Learning consistently outperformed 4 variations of ICA across different numbers of networks and noise levels (p$<$0.001). The NMF algorithms, which suppressed negative BOLD signal, had the poorest accuracy. Within each algorithm, encodings using sparser spatial networks (containing more zero-valued voxels) had higher classification accuracy (p$<$0.001). The success of sparse coding algorithms may suggest that algorithms which enforce sparse coding, discourage multitasking, and promote local specialization may capture better the underlying source processes than those which allow inexhaustible local processes such as ICA.

Bregman divergences play a central role in the design and analysis of a range of machine learning algorithms. This paper explores the use of Bregman divergences to establish reductions between such algorithms and their analyses. We present a new scaled isodistortion theorem involving Bregman divergences (scaled Bregman theorem for short) which shows that certain "Bregman distortions'" (employing a potentially non-convex generator) may be exactly re-written as a scaled Bregman divergence computed over transformed data. Admissible distortions include geodesic distances on curved manifolds and projections or gauge-normalisation, while admissible data include scalars, vectors and matrices. Our theorem allows one to leverage to the wealth and convenience of Bregman divergences when analysing algorithms relying on the aforementioned Bregman distortions. We illustrate this with three novel applications of our theorem: a reduction from multi-class density ratio to class-probability estimation, a new adaptive projection free yet norm-enforcing dual norm mirror descent algorithm, and a reduction from clustering on flat manifolds to clustering on curved manifolds. Experiments on each of these domains validate the analyses and suggest that the scaled Bregman theorem might be a worthy addition to the popular handful of Bregman divergence properties that have been pervasive in machine learning.

Solving l1 regularized optimization problems is common in the fields of computational biology, signal processing and machine learning. Such l1 regularization is utilized to find sparse minimizers of convex functions. A well-known example is the LASSO problem, where the l1 norm regularizes a quadratic function. A multilevel framework is presented for solving such l1 regularized sparse optimization problems efficiently. We take advantage of the expected sparseness of the solution, and create a hierarchy of problems of similar type, which is traversed in order to accelerate the optimization process. This framework is applied for solving two problems: (1) the sparse inverse covariance estimation problem, and (2) l1-regularized logistic regression. In the first problem, the inverse of an unknown covariance matrix of a multivariate normal distribution is estimated, under the assumption that it is sparse. To this end, an l1 regularized log-determinant optimization problem needs to be solved. This task is challenging especially for large-scale datasets, due to time and memory limitations. In the second problem, the l1-regularization is added to the logistic regression classification objective to reduce overfitting to the data and obtain a sparse model. Numerical experiments demonstrate the efficiency of the multilevel framework in accelerating existing iterative solvers for both of these problems.