But when it comes to AI, he doesn't mince words. AI is humanity's "biggest existential threat," he once proclaimed to some controversy. While that statement may be overblown, the fears aren't: AI will be the next technological force that transforms the face of society--for better or worse--much as the industrial revolution once did. The potential threats of AI are many, and most people agree that ethical AI that benefits humanity as a whole is critical for this technological quantum leap. But what exactly does "ethical AI" mean?

Artificial intelligence (AI) is transforming everything in our daily lives, from customer experience and healthcare to manufacturing and agriculture. In fact, in the UK alone investment for AI developers from venture capital increased more than 200 per cent last year. This comes as no surprise when you consider the growing number of AI startups being founded. According to a recent study from Stanford University, in the last 20 years there has been a 14-times increase in the number of AI startups. At the same time, we're seeing an increasing number of technology companies invest in AI development.

Artificial intelligence (AI) is transforming everything in our daily lives, from customer experience and healthcare to manufacturing and agriculture. In fact, in the UK alone investment for AI developers from venture capital increased more than 200 percent last year. This comes as no surprise when you consider the growing number of AI startups being founded. According to a recent study from Stanford University, in the last 20 years, there has been a 14-times increase in the number of AI startups.

Moojan started her career in the space of banking and corporate finance, before venturing into technology and start-ups. Not only is she the co-founder of Startup Sesame – an alliance of tech events in Europe, but she's also the founder of "Silk Road Startup too". Perhaps most importantly of all, Asghari is responsible for co-founding the Women in AI initiative – a group committed to closing the gender gap in the AI and ML fields. Moojan is helping to lead the way for innovative women everywhere. Devi Parikh is both an assistant professor for the school of Interactive computing for Georgia Tech, and research at Facebook AI research.

Machine learning algorithms can take important decisions, sometimes legally binding, about our everyday life. In most cases, however, these systems and decisions are neither regulated nor certified. Given the potential harm that these algorithms can cause, qualities such as fairness, accountability and transparency of predictive systems are of paramount importance. Recent literature suggested voluntary self-reporting on these aspects of predictive systems -- e.g., data sheets for data sets -- but their scope is often limited to a single component of a machine learning pipeline, and producing them requires manual labour. To resolve this impasse and ensure high-quality, fair, transparent and reliable machine learning systems, we developed an open source toolbox that can inspect selected fairness, accountability and transparency aspects of these systems to automatically and objectively report them back to their engineers and users. We describe design, scope and usage examples of this Python toolbox in this paper. The toolbox provides functionality for inspecting fairness, accountability and transparency of all aspects of the machine learning process: data (and their features), models and predictions. It is available to the public under the BSD 3-Clause open source licence.

Multilingual end-to-end (E2E) models have shown great promise in expansion of automatic speech recognition (ASR) coverage of the world's languages. They have shown improvement over monolingual systems, and have simplified training and serving by eliminating language-specific acoustic, pronunciation, and language models. This work presents an E2E multilingual system which is equipped to operate in low-latency interactive applications, as well as handle a key challenge of real world data: the imbalance in training data across languages. Using nine Indic languages, we compare a variety of techniques, and find that a combination of conditioning on a language vector and training language-specific adapter layers produces the best model. The resulting E2E multilingual model achieves a lower word error rate (WER) than both monolingual E2E models (eight of nine languages) and monolingual conventional systems (all nine languages). Index T erms: speech recognition, multilingual, RNN-T, residual adapter 1. Introduction Automatic speech recognition (ASR) systems that can transcribe speech in multiple languages, known as multilingual models, have gained popularity as an effective way to expand ASR coverage of the world's languages. Through shared learning of model elements across languages, they have been shown to outperform monolingual systems, particularly for those languages with less data.

We develop a general theory for the goodness-of-fit test to non-linear models. In particular, we assume that the observations are noisy samples of a sub-manifold defined by a non-linear map of some intrinsic structures. The observation noise is additive Gaussian. Our main result shows that the "residual" of the model fit, by solving a non-linear least-square problem, follows a (possibly non-central) $\chi^2$ distribution. The parameters of the $\chi^2$ distribution are related to the model order and dimension of the problem. The main result is established by making a novel connection between statistical test and differential geometry. We further present a method to select the model orders sequentially. We demonstrate the broad application of the general theory in a range of applications in machine learning and signal processing, including determining the rank of low-rank (possibly complex-valued) matrices and tensors, from noisy, partial, or indirect observations, determining the number of sources in signal demixing, and potential applications in determining the number of hidden nodes in neural networks.

Cryo-electron microscopy (cryo-EM) is a powerful technique for determining the structure of proteins and other macromolecular complexes at near-atomic resolution. In single particle cryo-EM, the central problem is to reconstruct the three-dimensional structure of a macromolecule from $10^{4-7}$ noisy and randomly oriented two-dimensional projections. However, the imaged protein complexes may exhibit structural variability, which complicates reconstruction and is typically addressed using discrete clustering approaches that fail to capture the full range of protein dynamics. Here, we introduce a novel method for cryo-EM reconstruction that extends naturally to modeling continuous generative factors of structural heterogeneity. This method encodes structures in Fourier space using coordinate-based deep neural networks, and trains these networks from unlabeled 2D cryo-EM images by combining exact inference over image orientation with variational inference for structural heterogeneity. We demonstrate that the proposed method, termed cryoDRGN, can perform ab initio reconstruction of 3D protein complexes from simulated and real 2D cryo-EM image data. To our knowledge, cryoDRGN is the first neural network-based approach for cryo-EM reconstruction and the first end-to-end method for directly reconstructing continuous ensembles of protein structures from cryo-EM images.

Propositionalization is the process of summarizing relational data into a tabular (attribute-value) format. The resulting table can next be used by any propositional learner. This approach makes it possible to apply a wide variety of learning methods to relational data. However, the transformation from relational to propositional format is generally not lossless: different relational structures may be mapped onto the same feature vector. At the same time, features may be introduced that are not needed for the learning task at hand. In general, it is hard to define a feature space that contains all and only those features that are needed for the learning task. This paper presents LazyBum, a system that can be considered a lazy version of the recently proposed OneBM method for propositionalization. LazyBum interleaves OneBM's feature construction method with a decision tree learner. This learner both uses and guides the propositionalization process. It indicates when and where to look for new features. This approach is similar to what has elsewhere been called dynamic propositionalization. In an experimental comparison with the original OneBM and with two other recently proposed propositionalization methods (nFOIL and MODL, which respectively perform dynamic and static propositionalization), LazyBum achieves a comparable accuracy with a lower execution time on most of the datasets.

Previous work has shown the unreliability of existing algorithms in the batch Reinforcement Learning setting, and proposed the theoretically-grounded Safe Policy Improvement with Baseline Bootstrapping (SPIBB) fix: reproduce the baseline policy in the uncertain state-action pairs, in order to control the variance on the trained policy performance. However, in many real-world applications such as dialogue systems, pharmaceutical tests or crop management, data is collected under human supervision and the baseline remains unknown. In this paper, we apply SPIBB algorithms with a baseline estimate built from the data. We formally show safe policy improvement guarantees over the true baseline even without direct access to it. Our empirical experiments on finite and continuous states tasks support the theoretical findings. It shows little loss of performance in comparison with SPIBB when the baseline policy is given, and more importantly, drastically and significantly outperforms competing algorithms both in safe policy improvement, and in average performance.