We developed a convolution neural network (CNN) on semi-regular triangulated meshes whose vertices have 6 neighbours. The key blocks of the proposed CNN, including convolution and down-sampling, are directly defined in a vertex domain. By exploiting the ordering property of semi-regular meshes, the convolution is defined on a vertex domain with strong motivation from the spatial definition of classic convolution. Moreover, the down-sampling of a semi-regular mesh embedded in a 3D Euclidean space can achieve a down-sampling rate of 4, 16, 64, etc. We demonstrated the use of this vertex-based graph CNN for the classification of mild cognitive impairment (MCI) and Alzheimer's disease (AD) based on 3169 MRI scans of the Alzheimer's Disease Neuroimaging Initiative (ADNI). We compared the performance of the vertex-based graph CNN with that of the spectral graph CNN.

Ranking is a core task in E-commerce recommender systems, which aims at providing an ordered list of items to users. Typically, a ranking function is learned from the labeled dataset to optimize the global performance, which produces a ranking score for each individual item. However, it may be sub-optimal because the scoring function applies to each item individually and does not explicitly consider the mutual influence between items, as well as the differences of users' preferences or intents. Therefore, we propose a personalized context-aware re-ranking model for E-commerce recommender systems. The proposed re-ranking model can be easily deployed as a follow-up modular after ranking by directly using the existing feature vectors of ranking. It directly optimizes the whole recommendation list by employing a transformer structure to efficiently encode the information of all items in the list. Specifically, the Transformer applies a self-attention mechanism that directly models the global relationships between any pair of items in the whole list. Besides, we introduce the personalized embedding to model the differences between feature distributions for different users. Experimental results on both offline benchmarks and real-world online E-commerce systems demonstrate the significant improvements of the proposed re-ranking model.

Reinforcement learning (RL) is a branch of machine learning which is employed to solve various sequential decision making problems without proper supervision. Due to the recent advancement of deep learning, the newly proposed Deep-RL algorithms have been able to perform extremely well in sophisticated high-dimensional environments. However, even after successes in many domains, one of the major challenge in these approaches is the high magnitude of interactions with the environment required for efficient decision making. Seeking inspiration from the brain, this problem can be solved by incorporating instance based learning by biasing the decision making on the memories of high rewarding experiences. This paper reviews various recent reinforcement learning methods which incorporate external memory to solve decision making and a survey of them is presented. We provide an overview of the different methods - along with their advantages and disadvantages, applications and the standard experimentation settings used for memory based models. This review hopes to be a helpful resource to provide key insight of the recent advances in the field and provide help in further future development of it.

Apposite Capital invests in Mirada Medical London, UK – 8th April 2019. Apposite Capital, the private equity firm with an exclusive focus on healthcare, today announces that it has invested £6m in preferred equity in Mirada Medical, a world leading medical imaging software company. The investment will support Mirada's development of artificial intelligence (#AI) based software for increasing the speed and accuracy of #cancer treatment planning applications, while delivering a more personalised level of care to patients. Apposite has invested alongside Mirada's founding investor, AlbionVC, and other private investors. Skymind raises $11.5M Skymind (*), a Y Combinator-incubated AI platform that aims to make deep learning more accessible to enterprises, today announced that it has raised an $11.5 million Series A round led by TransLink Capital, with participation from ServiceNow, Sumitomo's Presidio Ventures, UpHonest Capital and GovTech Fund.

The game of Go played between a DeepMind computer program and a human champion created an existential crisis of sorts for Marcus du Sautoy, a mathematician and professor at Oxford University. "I've always compared doing mathematics to playing the game of Go," he says, and Go is not supposed to be a game that a computer can easily play because it requires intuition and creativity. So when du Sautoy saw DeepMind's AlphaGo beat Lee Sedol, he thought that there had been a sea change in artificial intelligence that would impact other creative realms. He set out to investigate the role that AI can play in helping us understand creativity, and ended up writing The Creativity Code: Art and Innovation in the Age of AI (Harvard University Press). The Verge spoke to du Sautoy about different types of creativity, AI helping humans become more creative (instead of replacing them), and the creative fields where artificial intelligence struggles most.

A collaboration between researchers from China's Beihang University and Microsoft Research Asia has produced TableBank, a new image-based dataset for table detection and recognition built with novel weak supervision from Word and Latex documents on the Internet. Researchers built several strong baselines using SOTA models with deep neural networks, which will enable deployment of more deep learning methods to table detection and recognition tasks. TableBank has been open-sourced on Github. "Existing research for image-based table detection and recognition usually fine-tunes pre-trained models on out-of-domain data with a few thousands human labeled examples, which is difficult to generalize on real world applications. With TableBank that contains 417K high-quality labeled tables, we build several strong baselines using state-of-the-art models with deep neural networks."

Machine learning models come in many forms and cater to many use cases, and the proliferation of machine learning over the past decade has happened in three stages. First, there were breakthroughs in deep learning and the complementary emergence of cloud computing and big data. The second stage has been the creation and standardization of tools and frameworks for machine learning development. And most recently the deployment and in-life support of machine learning has come to the fore. But is this enough, or is there something missing?

Deep learning is at the core of recent spoken language understanding (SLU) related tasks. More precisely, deep neural networks (DNNs) drastically increased the performances of SLU systems, and numerous architectures have been proposed. In the real-life context of theme identification of telephone conversations, it is common to hold both a human, manual (TRS) and an automatically transcribed (ASR) versions of the conversations. Nonetheless, and due to production constraints, only the ASR transcripts are considered to build automatic classifiers. TRS transcripts are only used to measure the performances of ASR systems. Moreover, the recent performances in term of classification accuracy, obtained by DNN related systems are close to the performances reached by humans, and it becomes difficult to further increase the performances by only considering the ASR transcripts. This paper proposes to distillates the TRS knowledge available during the training phase within the ASR representation, by using a new generative adversarial network called M2H-GAN to generate a TRS-like version of an ASR document, to improve the theme identification performances.

We present a deep learning method for singing voice conversion. The proposed network is not conditioned on the text or on the notes, and it directly converts the audio of one singer to the voice of another. Training is performed without any form of supervision: no lyrics or any kind of phonetic features, no notes, and no matching samples between singers. The proposed network employs a single CNN encoder for all singers, a single WaveNet decoder, and a classifier that enforces the latent representation to be singer-agnostic. Each singer is represented by one embedding vector, which the decoder is conditioned on. In order to deal with relatively small datasets, we propose a new data augmentation scheme, as well as new training losses and protocols that are based on backtranslation. Our evaluation presents evidence that the conversion produces natural signing voices that are highly recognizable as the target singer.

Humor is a unique and creative communicative behavior displayed during social interactions. It is produced in a multimodal manner, through the usage of words (text), gestures (vision) and prosodic cues (acoustic). Understanding humor from these three modalities falls within boundaries of multimodal language; a recent research trend in natural language processing that models natural language as it happens in face-to-face communication. Although humor detection is an established research area in NLP, in a multimodal context it is an understudied area. This paper presents a diverse multimodal dataset, called UR-FUNNY, to open the door to understanding multimodal language used in expressing humor. The dataset and accompanying studies, present a framework in multimodal humor detection for the natural language processing community. UR-FUNNY is publicly available for research.