During Apple's fourth-quarter earnings call with analysts, CEO Tim Cook said, "AR is going to change everything." Augmented reality (AR) is shaping an entirely new paradigm for mass technology use. We've quickly evolved from typing on our PC keyboards, to the point-and-click of the mouse, to the smartphone's tap or swipe, to simply asking Alexa or Siri to do things for us. Now AR brings us to the age of holographic computing. Along with animojies and Pokémon and face filters, a fresh and futuristic user interface is emerging.

In this paper, we propose a new optimization algorithm for sparse logistic regression based on a stochastic version of the Douglas-Rachford splitting method. Our algorithm sweeps the training set by randomly selecting a mini-batch of data at each iteration, and it allows us to update the variables in a block coordinate manner. Our approach leverages the proximity operator of the logistic loss, which is expressed with the generalized Lambert W function. Experiments carried out on standard datasets demonstrate the efficiency of our approach w.r.t. stochastic gradient-like methods.

Over the last few years, deep learning has revolutionized the field of machine learning by dramatically improving the state-of-the-art in various domains. However, as the size of supervised artificial neural networks grows, typically so does the need for larger labeled datasets. Recently, crowdsourcing has established itself as an efficient and cost-effective solution for labeling large sets of data in a scalable manner, but it often requires aggregating labels from multiple noisy contributors with different levels of expertise. In this paper, we address the problem of learning deep neural networks from crowds. We begin by describing an EM algorithm for jointly learning the parameters of the network and the reliabilities of the annotators. Then, a novel general-purpose crowd layer is proposed, which allows us to train deep neural networks end-to-end, directly from the noisy labels of multiple annotators, using only backpropagation. We empirically show that the proposed approach is able to internally capture the reliability and biases of different annotators and achieve new state-of-the-art results for various crowdsourced datasets across different settings, namely classification, regression and sequence labeling.

Characterising tractable fragments of the constraint satisfaction problem (CSP) is an important challenge in theoretical computer science and artificial intelligence. Forbidding patterns (generic sub-instances) provides a means of defining CSP fragments which are neither exclusively language-based nor exclusively structure-based. It is known that the class of binary CSP instances in which the broken-triangle pattern (BTP) does not occur, a class which includes all tree-structured instances, are decided by arc consistency (AC), a ubiquitous reduction operation in constraint solvers. We provide a characterisation of simple partially-ordered forbidden patterns which have this AC-solvability property. It turns out that BTP is just one of five such AC-solvable patterns. The four other patterns allow us to exhibit new tractable classes.

File Photo: People attend an anti-suicide event in Vilnius, Lithuania, Sept. 22, 2015. SAN FRANCISCO, Nov. 27 (Xinhua) -- U.S. hi-tech giant Facebook said Monday it is using artificial intelligence (AI) technology, including pattern recognition, to detect whether someone is expressing thoughts of suicide in a post or live video. "Facebook is a place where friends and family are already connected and we are able to help connect a person in distress with people who can support them," the company said in a press release. It said it is "using pattern recognition to detect posts or live videos where someone might be expressing thoughts of suicide, and to help respond to reports faster." The world's largest social network also said it now has more workers to review reports of suicide and self-harm.

Ginni Rometty, CEO of IBM, creators of the Watson AI system, spoke of AI and individual interactivity, and the fear individuals have of their positions being rendered obsolete. AI can sort potatoes into those set for French fry production, or those better suited to crisp or potato wedge products, meaning less waste. AI technology is being developed that could render fast food burger cooks obsolete through new and innovative cooking methods, entirely automated. Artificial intelligence (AI) and its impact on business was a key talking point at this year's World Economic Forum in Davos, Switzerland (Davos 2017). Speaking at an AI panel at Davos 2017, Microsoft CEO Satya Nadella discussed how simple it was to eliminate human input altogether: "its augmentation or replacement, that's a design choice. You can say replacement [of humans] is the goal, or you can say augmentation is the goal."

Can robots and humans make babies together? This is a serious question inspired by some of the advances already achieved in the 21st century by researchers in cell biology and in a discipline variously known as biorobotics, synthetic biology, or bionanotechnology. Although it had long been a truth universally acknowledged that sexual intercourse was an essential precursor to conception, it was only around 150 years ago that early studies of embryology revealed the reason why, according to the dogma of the time, intercourse was "essential" in human reproduction. The reason was that only an egg from a female, fertilized by a sperm from a male, can result in a live birth. But thanks to the Nobel prize winning work of researchers like embryologist John Gurdon and stem cell researcher Shinya Yamanaka, it has become possible during the past few years to create both sperm cells and eggs in the laboratory from skin cells, obviating the need for a human mother or father to kick off the reproductive process.

t-distributed Stochastic Neighborhood Embedding (t-SNE) is a method for dimensionality reduction and visualization that has become widely popular in recent years. Efficient implementations of t-SNE are available, but they scale poorly to datasets with hundreds of thousands to millions of high dimensional data-points. We present Fast Fourier Transform-accelerated Interpolation-based t-SNE (FIt-SNE), which dramatically accelerates the computation of t-SNE. The most time-consuming step of t-SNE is a convolution that we accelerate by interpolating onto an equispaced grid and subsequently using the fast Fourier transform to perform the convolution. We also optimize the computation of input similarities in high dimensions using multi-threaded approximate nearest neighbors. We further present a modification to t-SNE called "late exaggeration," which allows for easier identification of clusters in t-SNE embeddings. Finally, for datasets that cannot be loaded into the memory, we present out-of-core randomized principal component analysis (oocPCA), so that the top principal components of a dataset can be computed without ever fully loading the matrix, hence allowing for t-SNE of large datasets to be computed on resource-limited machines.

Artificial neural network (ANN) is a very useful tool in solving learning problems. Boosting the performances of ANN can be mainly concluded from two aspects: optimizing the architecture of ANN and normalizing the raw data for ANN. In this paper, a novel method which improves the effects of ANN by preprocessing the raw data is proposed. It totally leverages the fact that different features should play different roles. The raw data set is firstly preprocessed by principle component analysis (PCA), and then its principle components are weighted by their corresponding eigenvalues. Several aspects of analysis are carried out to analyze its theory and the applicable occasions. Three classification problems are launched by an active learning algorithm to verify the proposed method. From the empirical results, conclusion comes to the fact that the proposed method can significantly improve the performance of ANN.

I'm an AI researcher (based outside Oxford UK) at eXvisory.ai. It's an AI web chat application that guides consumers through finding and fixing problems with their mobile phones or tablets. Instead of chatting with a scripted human support engineer consumers chat with a scripted AI. Machine learning AI is amazing for matching single questions to single answers (given lots of high quality Q&A training data) but not so hot with conversational or back-and-forth Q&A. It's not an algorithmic problem -- it's just much harder to obtain high quality conversational Q&A training data.