When deploying machine learning models in high-stakes real-world environments such as health care, it is crucial to accurately assess the uncertainty concerning a model's prediction on abnormal inputs. However, there is a scarcity of literature analyzing this problem on medical data, especially on mixed-type tabular data such as Electronic Health Records. We close this gap by presenting a series of tests including a large variety of contemporary uncertainty estimation techniques, in order to determine whether they are able to identify out-of-distribution (OOD) patients. In contrast to previous work, we design tests on realistic and clinically relevant OOD groups, and run experiments on real-world medical data. We find that almost all techniques fail to achieve convincing results, partly disagreeing with earlier findings.

Due to accessible big data collections from consumers, products, and stores, advanced sales forecasting capabilities have drawn great attention from many companies especially in the retail business because of its importance in decision making. Improvement of the forecasting accuracy, even by a small percentage, may have a substantial impact on companies' production and financial planning, marketing strategies, inventory controls, supply chain management, and eventually stock prices. Specifically, our research goal is to forecast the sales of each product in each store in the near future. Motivated by tensor factorization methodologies for personalized context-aware recommender systems, we propose a novel approach called the Advanced Temporal Latent-factor Approach to Sales forecasting (ATLAS), which achieves accurate and individualized prediction for sales by building a single tensor-factorization model across multiple stores and products. Our contribution is a combination of: tensor framework (to leverage information across stores and products), a new regularization function (to incorporate demand dynamics), and extrapolation of tensor into future time periods using state-of-the-art statistical (seasonal auto-regressive integrated moving-average models) and machine-learning (recurrent neural networks) models. The advantages of ATLAS are demonstrated on eight product category datasets collected by the Information Resource, Inc., where a total of 165 million weekly sales transactions from more than 1,500 grocery stores over 15,560 products are analyzed.

We study the effect of mini-batching on the loss landscape of deep neural networks using spiked, field-dependent random matrix theory. We demonstrate that the magnitude of the extremal values of the batch Hessian are larger than those of the empirical Hessian. We also derive similar results for the Generalised Gauss-Newton matrix approximation of the Hessian. As a consequence of our theorems we derive an analytical expressions for the maximal learning rates as a function of batch size, informing practical optimisation schemes for both stochastic gradient descent (linear scaling) and adaptive algorithms such as Adam (square root scaling). Whilst the linear scaling for stochastic gradient descent has been derived under more restrictive conditions, which we generalise, the square root scaling rule for adaptive optimisers is, to our knowledge, completely novel. For stochastic Second-order methods and adaptive methods, we derive that the minimal damping coefficient is proportional to the ratio of the learning rate to batch size.

Neural link predictors are immensely useful for identifying missing edges in large scale Knowledge Graphs. However, it is still not clear how to use these models for answering more complex queries that arise in a number of domains, such as queries using logical conjunctions, disjunctions, and existential quantifiers, while accounting for missing edges. In this work, we propose a framework for efficiently answering complex queries on incomplete Knowledge Graphs. We translate each query into an end-to-end differentiable objective, where the truth value of each atom is computed by a pre-trained neural link predictor. We then analyse two solutions to the optimisation problem, including gradient-based and combinatorial search. In our experiments, the proposed approach produces more accurate results than state-of-the-art methods -- black-box neural models trained on millions of generated queries -- without the need of training on a large and diverse set of complex queries. Using orders of magnitude less training data, we obtain relative improvements ranging from 8% up to 40% in Hits@3 across different knowledge graphs containing factual information. Finally, we demonstrate that it is possible to explain the outcome of our model in terms of the intermediate solutions identified for each of the complex query atoms.

Modern industrial applications require robots to be able to operate in unpredictable environments, and programs to be created with a minimal effort, as there may be frequent changes to the task. In this paper, we show that genetic programming can be effectively used to learn the structure of a behavior tree (BT) to solve a robotic task in an unpredictable environment. Moreover, we propose to use a simple simulator for the learning and demonstrate that the learned BTs can solve the same task in a realistic simulator, reaching convergence without the need for task specific heuristics. The learned solution is tolerant to faults, making our method appealing for real robotic applications.

Research in NLP lacks geographic diversity, and the question of how NLP can be scaled to low-resourced languages has not yet been adequately solved. "Low-resourced"-ness is a complex problem going beyond data availability and reflects systemic problems in society. In this paper, we focus on the task of Machine Translation (MT), that plays a crucial role for information accessibility and communication worldwide. Despite immense improvements in MT over the past decade, MT is centered around a few high-resourced languages. As MT researchers cannot solve the problem of low-resourcedness alone, we propose participatory research as a means to involve all necessary agents required in the MT development process. We demonstrate the feasibility and scalability of participatory research with a case study on MT for African languages. Its implementation leads to a collection of novel translation datasets, MT benchmarks for over 30 languages, with human evaluations for a third of them, and enables participants without formal training to make a unique scientific contribution. Benchmarks, models, data, code, and evaluation results are released under https://github.com/masakhane-io/masakhane-mt.

The agreements followed the announcement of Saudi Arabia's National Strategy for Data and Artificial Intelligence, launched during the Global AI Summit Saudi Arabia has signed a series of partnership agreements with international tech companies to advance artifical intelligence (AI) in the kingdom. The agreements, which were rigned at the virtual Global AI Summit held in Riyadh, are underpinned by Saudi Arabia's newly-launched National Strategy for Data and Artificial Intelligence (NSDAI). Saudi Arabia's National Center for Artificial Intelligence (NCAI) announced a memorandum of understanding (MoU) with China's Huawei to enable strategic cooperation on the kingdom's National AI Capability Development Program. Under the MoU, Huawei will support the NCAI to train Saudi AI engineers and students, and to address Arabic language AI-related capabilities. NCAI and Huawei will also explore the creation of an AI Capability Platform to localise technology solutions.

Japanese researchers envision a future where swarms of cyborg cockroaches roam freely inside homes, carrying out a variety of small tasks. A team at the University of Tsukuba modified Madagascar cockroaches with cybernetic implants that navigate the insects up walls and across floors – places other robots have difficult accessing. Called'Calmbots,' the cockroaches were installed with electrodes, a chip antenna, battery and a pixel strapped to its back that can be used as a display. Researchers say the cyborgs can transport objects around the home, drawing things on paper and may one day act as an'input or haptic interfaces or an audio device. Calmbots are a project of Digital Nature Group, a department at the university, which aims to release their creations into people's homes.

Early this spring as the pandemic began accelerating, AJ Venkatakrishnan took genetic data from 10,967 samples of the novel coronavirus and fed it into a machine. The Stanford-trained data scientist did not have a particular hypothesis, but he was hoping the artificial intelligence would pinpoint possible weaknesses that could be exploited to develop therapies. He was awed when the program reported back that the new virus appeared to have a snippet of DNA code - "RRARSAS" - distinct from its predecessor coronaviruses. This sequence, he learned, mimics a protein that helps the human body regulate salt and fluid balance. Venkatakrishnan, director of scientific research and partnerships at AI start-up Nference, wondered whether this change might allow the virus to act as a kind of Trojan horse. Could this explain its high infection and transmission rates?

One of the main barriers to the broader adoption of algorithmic fairness in machine learning is the trade-off between fairness and performance of ML models: many practitioners are unwilling to sacrifice the performance of their ML model for fairness. In this paper, we show that this trade-off may not be necessary. If the algorithmic biases in an ML model are due to sampling biases in the training data, then enforcing algorithmic fairness may improve the performance of the ML model on unbiased test data. We study conditions under which enforcing algorithmic fairness helps practitioners learn the Bayes decision rule for (unbiased) test data from biased training data. We also demonstrate the practical implications of our theoretical results in real-world ML tasks.