Research over the last five years shows a clear improvement in computer-aided detection (CAD), specifically in disease prediction from medical images [36, 61, 105, 130, 134] as well as from Electronic Health Records (EHR) [113], by using deep neural networks (DNN) and treating the problem as supervised classification or segmentation tasks. Recently, Topol [129] indicates that the need for diagnosis and reporting from image-based examinations far exceeds the current medical capacity of physicians in the US. This situation promotes the development of automatic image-based diagnosis as well as automatic reporting. Furthermore, the lack of specialist physicians is even more critical in resource-limited countries [111], and therefore the expected impacts of this technology would become even more relevant. However, the elaboration of high-quality medical reports from medical images, such as chest X-rays, computed tomography (CT) or magnetic resonance (MRI) scans, is a task that requires a trained radiologist with years of experience.

Tucker decomposition is one of the most popular models for analyzing and compressing large-scale tensorial data. Existing Tucker decomposition algorithms usually rely on a single solver to compute the factor matrices and core tensor, and are not flexible enough to adapt with the diversities of the input data and the hardware. Moreover, to exploit highly efficient GEMM kernels, most Tucker decomposition implementations make use of explicit matricizations, which could introduce extra costs in terms of data conversion and memory usage. In this paper, we present a-Tucker, a new framework for input-adaptive and matricization-free Tucker decomposition of dense tensors. A mode-wise flexible Tucker decomposition algorithm is proposed to enable the switch of different solvers for the factor matrices and core tensor, and a machine-learning adaptive solver selector is applied to automatically cope with the variations of both the input data and the hardware. To further improve the performance and enhance the memory efficiency, we implement a-Tucker in a fully matricization-free manner without any conversion between tensors and matrices. Experiments with a variety of synthetic and real-world tensors show that a-Tucker can substantially outperform existing works on both CPUs and GPUs.

The news: Facebook is open-sourcing a new AI language model called M2M-100 that can translate between any pair among 100 languages. This is in contrast to previous multilingual models, which heavily rely on English as an intermediate. A Chinese to French translation, for example, typically passes from Chinese to English and then English to French, which increases the chance of introducing errors. Data curation: The model was trained on 7.5 billion sentence pairs. In order to compile a data set that large, the researchers relied heavily on automated curation.

A major new report led by the Novartis Foundation and Microsoft shows how investment in data and AI is critical to drive the health system improvements needed to respond to and recover from the COVID-19 pandemic and the world's other greatest healthcare challenges. Reimagining Global Health through Artificial Intelligence: The Roadmap to AI Maturity was developed by the Broadband Commission Working Group on Digital and AI in Health, which the Novartis Foundation and Microsoft co-chair. Based on a landscape review of over 300 existing use cases of AI in health, the report shows how AI is already disrupting health and care. It then presents a roadmap to help countries use AI to transform their health systems from being reactive to proactive, predictive, and even preventive. Low- and middle-income countries (LMIC) that grapple with systemic health challenges such as a shortage of health workers, underserved populations, rapid urbanization and disinformation have the most to gain from AI – but they also have the most to lose.

The work in this repo was presented and demoed at the 2019 Experimental A.I. in Games (EXAG) workshop, an AIIDE workshop. Feel free to join the conversation surrounding this work via my Twitter, and on r/MachineLearning. The player builds places urban structures on a 2D map. In certain configurations, these structures invite population and vertical development. Reinforcement Learning agents are rewarded as a function of population or other city-wide metrics.

Rapid advances in technology, connectivity and telecommunications are conspiring to make Africa's large, rapidly growing population a valuable asset for the automation revolution. It is imperative that Africa quickly develop agency in data and artificial intelligence and it will be lucrative for investors who support them by financing Africa's telecom and data backbone. Africa must urgently develop cogent digital strategy. This at first seems fanciful, or even superfluous, given the continent's relative lack of more basic development. Indeed, there are myriad other challenges to which most would assign primacy.

The pace of digital transformation has notably picked up in the past decade, as enterprises invest in technology to retain their competitive edge and avoid having their market share eroded by disruptive newcomers. Organisations' ability to out-innovate their competitors in this way often requires a full-scale modernisation of the IT infrastructure stack underpinning their operations so they are better positioned to respond to the changing needs of their customers. For many enterprises, this process of modernisation has seen them look to invest in making their private, virtualised datacentres and server rooms more agile, responsive and easier to manage by investing in software-defined networking (SDN) technologies and automation tools. Such investments can help enterprises make better and more efficient use of their existing compute capacity, but that alone may not be enough to stave off competitive threats, prompting some IT leaders to weigh up a move to the public cloud. The benefits of such an approach are well-documented and proven, with the public cloud offering enterprises ready access to an almost infinite supply of cloud-based compute resources that can be set to auto-scale in line with peaks and troughs in demand, meaning enterprises only pay for what they use.

Graph neural network models have been extensively used to learn node representations for graph structured data in an end-to-end setting. These models often rely on localized first order approximations of spectral graph convolutions and hence are unable to capture higher-order relational information between nodes. Probabilistic Graphical Models form another class of models that provide rich flexibility in incorporating such relational information but are limited by inefficient approximate inference algorithms at higher order. In this paper, we propose to combine these approaches to learn better node and graph representations. First, we derive an efficient approximate sum-product loopy belief propagation inference algorithm for higher-order PGMs. We then embed the message passing updates into a neural network to provide the inductive bias of the inference algorithm in end-to-end learning. This gives us a model that is flexible enough to accommodate domain knowledge while maintaining the computational advantage. We further propose methods for constructing higher-order factors that are conditioned on node and edge features and share parameters wherever necessary. Our experimental evaluation shows that our model indeed captures higher-order information, substantially outperforming state-of-the-art $k$-order graph neural networks in molecular datasets.

We present an information-based uncertainty quantification method for general Markov Random Fields. Markov Random Fields (MRFs) are structured, probabilistic graphical models over undirected graphs, and provide a fundamental unifying modeling tool for statistical mechanics, probabilistic machine learning, and artificial intelligence. Typically MRFs are complex and high-dimensional with nodes and edges (connections) built in a modular fashion from simpler, low-dimensional probabilistic models and their local connections; in turn, this modularity allows to incorporate available data to MRFs and efficiently simulate them by leveraging their graph-theoretic structure. Learning graphical models from data and/or constructing them from physical modeling and constraints necessarily involves uncertainties inherited from data, modeling choices, or numerical approximations. These uncertainties in the MRF can be manifested either in the graph structure or the probability distribution functions, and necessarily will propagate in predictions for quantities of interest. Here we quantify such uncertainties using tight, information-based bounds on the predictions of quantities of interest; these bounds take advantage of the graphical structure of MRFs and are capable of handling the inherent high-dimensionality of such graphical models. We demonstrate our methods in MRFs for medical diagnostics and statistical mechanics models. In the latter, we develop uncertainty quantification bounds for finite-size effects and phase diagrams, which constitute two of the typical predictions goals of statistical mechanics modeling.

This paper presents and characterizes an Open Application Repository for Federated Learning (OARF), a benchmark suite for federated machine learning systems. Previously available benchmarks for federated learning have focused mainly on synthetic datasets and use a very limited number of applications. OARF includes different data partitioning methods (horizontal, vertical and hybrid) as well as emerging applications in image, text and structured data, which represent different scenarios in federated learning. Our characterization shows that the benchmark suite is diverse in data size, distribution, feature distribution and learning task complexity. We have developed reference implementations, and evaluated the important aspects of federated learning, including model accuracy, communication cost, differential privacy, secure multiparty computation and vertical federated learning.