Transformer benefits from the high parallelization of attention networks in fast training, but it still suffers from slow decoding partially due to the linear dependency O(m) of the decoder self-attention on previous target words at inference. In this paper, we propose a generalized average attention network (AAN+) aiming at speeding up decoding by reducing the dependency from O(m) to O(1). We find that the learned self-attention weights in the decoder follow some patterns which can be approximated via a dynamic structure. Based on this insight, we develop AAN+, extending our previously proposed average attention (Zhang et al., 2018a, AAN) to support more general position- and content-based attention patterns. AAN+ only requires to maintain a small constant number of hidden states during decoding, ensuring its O(1) dependency. We apply AAN+ as a drop-in replacement of the decoder selfattention and conduct experiments on machine translation (with diverse language pairs), table-to-text generation and document summarization. With masking tricks and dynamic programming, AAN+ enables Transformer to decode sentences around 20% faster without largely compromising in the training speed and the generation performance. Our results further reveal the importance of the localness (neighboring words) in AAN+ and its capability in modeling long-range dependency.

After turning heads with her paintings of icons like Queen Elizabeth II and Sir Paul McCartney, artificial intelligence robot Ai-Da made a historic appearance in the UK's House of Lords on October 11, where the sophisticated automaton answered questions as part of a wider inquiry into the relationship between technology and creativity in the modern world. Ai-Da's parliamentary cameo is a reminder that emerging AI technologies have a more diverse range of positive applications than many people might think. As well as its extensive use in the creative sector (which is often heralded as the defining distinction between human and robot), AI also has much to offer in terms of improving safety, efficiency, and the overall user experience in a variety of different industries and areas. Ai-Da might be commanding much of the headlines surrounding AI art at the moment, but she represents the merest tip of the iceberg when it comes to AI's contribution to the creative sector. Indeed, the gaming industry has been one of the driving forces and guinea pigs behind the development of the technology, with the first chess algorithms coming out in the 1950s and IBM's Deep Blue creating quite the storm in 1997 when it beat then-Grandmaster Garry Kasparov.

A gaggle of new AI projects are coming soon from Google, including disaster monitoring tools and a service that uses machine intelligence to generate custom videos. The company announced the array of initiatives at its AI@ event this week. The most practical development: Google is expanding its AI-powered disaster tracking and response systems. The company rolled out a wildfire tracking tool during the apocalyptic 2020 fire season. The tool aims to track wildfire movements in real time using satellite imagery, on-the-ground data, and AI predictions.

Join us on November 9 to learn how to successfully innovate and achieve efficiency by upskilling and scaling citizen developers at the Low-Code/No-Code Summit. From COVID--19 to monkeypox and intermittent polio scares, concerns around public health crises have significantly increased over the past several years. Living in a globally connected world amidst climate change and a growing population has enabled the emergence of more frequent viruses and fostered their spread. A research study last year estimated that the probability of novel disease outbreaks will grow three-fold in the next few decades. Fortunately, there have been significant technological developments that can help minimize the impact of these global health issues.

Australian surf lifesavers are increasingly using drones to spot sharks at the beach before they get too close to swimmers. But just how reliable are they? Discerning whether that dark splodge in the water is a shark or just, say, seaweed isn't always straightforward and, in reasonable conditions, drone pilots generally make the right call only 60% of the time. While this has implications for public safety, it can also lead to unnecessary beach closures and public alarm. Engineers are trying to boost the accuracy of these shark-spotting drones with artificial intelligence (AI).

Artificial intelligence is experiencing exponential growth and is being used by thousands of businesses worldwide. It is easing our daily lives and offering solutions to the most challenging issues. Let's look at some of the most cutting-edge AI startups established in Finland. Although digital or online learning is developing quickly, it still has many shortcomings, including a lack of simplicity and personalization. Claned is a personalized online learning platform revolutionizing the digital learning arena.

In natural language processing (NLP), the context of a word or sentence plays an essential role. Contextual information such as the semantic representation of a passage or historical dialogue forms an essential part of a conversation and a precise understanding of the present phrase or sentence. However, the standard attention mechanisms typically generate weights using query and key but ignore context, forming a Bi-Attention framework, despite their great success in modeling sequence alignment. This Bi-Attention mechanism does not explicitly model the interactions between the contexts, queries and keys of target sequences, missing important contextual information and resulting in poor attention performance. Accordingly, a novel and general triple-attention (Tri-Attention) framework expands the standard Bi-Attention mechanism and explicitly interacts query, key, and context by incorporating context as the third dimension in calculating relevance scores. Four variants of Tri-Attention are generated by expanding the two-dimensional vector-based additive, dot-product, scaled dot-product, and bilinear operations in Bi-Attention to the tensor operations for Tri-Attention. Extensive experiments on three NLP tasks demonstrate that Tri-Attention outperforms about 30 state-of-the-art non-attention, standard Bi-Attention, contextual Bi-Attention approaches and pretrained neural language models1.

Purpose: Hard-to-interpret Black-box Machine Learning (ML) were often used for early Alzheimer's Disease (AD) detection. Methods: To interpret eXtreme Gradient Boosting (XGBoost), Random Forest (RF), and Support Vector Machine (SVM) black-box models a workflow based on Shapley values was developed. All models were trained on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset and evaluated for an independent ADNI test set, as well as the external Australian Imaging and Lifestyle flagship study of Ageing (AIBL), and Open Access Series of Imaging Studies (OASIS) datasets. Shapley values were compared to intuitively interpretable Decision Trees (DTs), and Logistic Regression (LR), as well as natural and permutation feature importances. To avoid the reduction of the explanation validity caused by correlated features, forward selection and aspect consolidation were implemented. Results: Some black-box models outperformed DTs and LR. The forward-selected features correspond to brain areas previously associated with AD. Shapley values identified biologically plausible associations with moderate to strong correlations with feature importances. The most important RF features to predict AD conversion were the volume of the amygdalae, and a cognitive test score. Good cognitive test performances and large brain volumes decreased the AD risk. The models trained using cognitive test scores significantly outperformed brain volumetric models ($p<0.05$). Cognitive Normal (CN) vs. AD models were successfully transferred to external datasets. Conclusion: In comparison to previous work, improved performances for ADNI and AIBL were achieved for CN vs. Mild Cognitive Impairment (MCI) classification using brain volumes. The Shapley values and the feature importances showed moderate to strong correlations.

Single-cell technologies are revolutionizing the entire field of biology. The large volumes of data generated by single-cell technologies are high-dimensional, sparse, heterogeneous, and have complicated dependency structures, making analyses using conventional machine learning approaches challenging and impractical. In tackling these challenges, deep learning often demonstrates superior performance compared to traditional machine learning methods. In this work, we give a comprehensive survey on deep learning in single-cell analysis. We first introduce background on single-cell technologies and their development, as well as fundamental concepts of deep learning including the most popular deep architectures. We present an overview of the single-cell analytic pipeline pursued in research applications while noting divergences due to data sources or specific applications. We then review seven popular tasks spanning through different stages of the single-cell analysis pipeline, including multimodal integration, imputation, clustering, spatial domain identification, cell-type deconvolution, cell segmentation, and cell-type annotation. Under each task, we describe the most recent developments in classical and deep learning methods and discuss their advantages and disadvantages. Deep learning tools and benchmark datasets are also summarized for each task. Finally, we discuss the future directions and the most recent challenges. This survey will serve as a reference for biologists and computer scientists, encouraging collaborations.

Learning curves provide insight into the dependence of a learner's generalization performance on the training set size. This important tool can be used for model selection, to predict the effect of more training data, and to reduce the computational complexity of model training and hyperparameter tuning. This review recounts the origins of the term, provides a formal definition of the learning curve, and briefly covers basics such as its estimation. Our main contribution is a comprehensive overview of the literature regarding the shape of learning curves. We discuss empirical and theoretical evidence that supports well-behaved curves that often have the shape of a power law or an exponential. We consider the learning curves of Gaussian processes, the complex shapes they can display, and the factors influencing them. We draw specific attention to examples of learning curves that are ill-behaved, showing worse learning performance with more training data. To wrap up, we point out various open problems that warrant deeper empirical and theoretical investigation. All in all, our review underscores that learning curves are surprisingly diverse and no universal model can be identified.