It is not easy to build a machine learning model. It is even harder to deploy a service in production. But even if you managed to stick all the pipelines together, things do not stop here. Once the model is in use, we immediately have to think about operating it smoothly. It is now delivering the business value, after all!

My friend Mike Carey, who has died aged 71, was a pioneer in the development of speech recognition and digital audio, including digital audio broadcasting (DAB). After a formative period at the Post Office research labs, Keele University and Mitel Telecom, he left in 1985 with Adrian Anderson and me to establish Ensigma, a company that quickly established itself as a leader in the research and development of digital speech and audio applications, with uses in mobile telephony, broadcasting and automated speech recognition systems. Mike was born in Preston, to Ethel (nee Glover), a weaver and housewife, and Stephen, a miner-cum-decorator. At Preston Catholic college, a local grammar, he met Elizabeth Mercer at a school dance, for once beating the cross-country champion across the floor. They married in 1969, by which time Mike was on a Post Office scholarship to study electrical engineering at Imperial College, London, supported throughout by extra work taken on by his parents.

Having one of these in your resume can make a lot of difference. Data science is one of the most versatile fields ever around; even its name is not very explanatory of what the field actually involves. Perhaps that's one reason people find this field quite challenging and difficult to get into and even more difficult to show professionalism. It is well known within the data science community that to be a "good" data scientist is all about how strong of a portfolio you build, how diverse your projects are, and how well they show your ability to solve any problem creatively and efficiently. Although being a data scientist -- or have a specialty in any of its branches -- doesn't require a university degree, having some certificate that proves your profession in some aspects of the field can transform your portfolio and take your career on step further.

During the spring semester of my junior year in college, I had the opportunity to study abroad in Copenhagen, Denmark. I had never been to Europe before that, so I was incredibly excited to immerse myself into a new culture, meet new people, travel to new places, and, most important, encounter a new language. Now although English is not my native language (Vietnamese is), I have learned and spoken it since early childhood, making it second-nature. Danish, on the other hand, is an incredibly complicated language with a very different sentence and grammatical structure. Before my trip, I tried to learn a bit of Danish using the app Duolingo; however, I only got a hold of simple phrases such as Hello (Hej) and Good Morning (God Morgen).

Deep learning is a type of machine learning, which is a subset of artificial intelligence. Machine learning is about computers being able to think and act with less human intervention; deep learning is about computers learning to think using structures modeled on the human brain. Machine learning requires less computing power; deep learning typically needs less ongoing human intervention. Deep learning can analyze images, videos, and unstructured data in ways machine learning can't easily do. Every industry will have career paths that involve machine and deep learning.

The development of artificial intelligence (AI) has made various industries eager to explore the benefits of AI. There is an increasing amount of research surrounding AI, most of which is centred on the development of new AI algorithms and techniques. However, the advent of AI is bringing an increasing set of practical problems related to AI model lifecycle management that need to be investigated. We address this gap by conducting a systematic mapping study on the lifecycle of AI model. Through quantitative research, we provide an overview of the field, identify research opportunities, and provide suggestions for future research. Our study yields 405 publications published from 2005 to 2020, mapped in 5 different main research topics, and 31 sub-topics. We observe that only a minority of publications focus on data management and model production problems, and that more studies should address the AI lifecycle from a holistic perspective.

We propose a distributionally robust support vector machine with a fairness constraint that encourages the classifier to be fair in view of the equality of opportunity criterion. We use a type-$\infty$ Wasserstein ambiguity set centered at the empirical distribution to model distributional uncertainty and derive an exact reformulation for worst-case unfairness measure. We establish that the model is equivalent to a mixed-binary optimization problem, which can be solved by standard off-the-shelf solvers. We further prove that the expectation of the hinge loss objective function constitutes an upper bound on the misclassification probability. Finally, we numerically demonstrate that our proposed approach improves fairness with negligible loss of predictive accuracy.

Standard Neural Networks can learn mathematical operations, but they do not extrapolate. Extrapolation means that the model can apply to larger numbers, well beyond those observed during training. Recent architectures tackle arithmetic operations and can extrapolate; however, the equally important problem of quantitative reasoning remains unaddressed. In this work, we propose a novel architectural element, the Neural Status Register (NSR), for quantitative reasoning over numbers. Our NSR relaxes the discrete bit logic of physical status registers to continuous numbers and allows end-to-end learning with gradient descent. Experiments show that the NSR achieves solutions that extrapolate to numbers many orders of magnitude larger than those in the training set. We successfully train the NSR on number comparisons, piecewise discontinuous functions, counting in sequences, recurrently finding minimums, finding shortest paths in graphs, and comparing digits in images.

Reinforcement learning (RL) algorithms are typically limited to learning a single solution of a specified task, even though there often exists diverse solutions to a given task. Compared with learning a single solution, learning a set of diverse solutions is beneficial because diverse solutions enable robust few-shot adaptation and allow the user to select a preferred solution. Although previous studies have showed that diverse behaviors can be modeled with a policy conditioned on latent variables, an approach for modeling an infinite set of diverse solutions with continuous latent variables has not been investigated. In this study, we propose an RL method that can learn infinitely many solutions by training a policy conditioned on a continuous or discrete low-dimensional latent variable. Through continuous control tasks, we demonstrate that our method can learn diverse solutions in a data-efficient manner and that the solutions can be used for few-shot adaptation to solve unseen tasks.

Classical network embeddings create a low dimensional representation of the learned relationships between features across nodes. Such embeddings are important for tasks such as link prediction and node classification. In the current paper, we consider low dimensional embeddings of dynamic networks, that is a family of time varying networks where there exist both temporal and spatial link relationships between nodes. We present novel embedding methods for a dynamic network based on higher order tensor decompositions for tensorial representations of the dynamic network. In one sense, our embeddings are analogous to spectral embedding methods for static networks. We provide a rationale for our algorithms via a mathematical analysis of some potential reasons for their effectiveness. Finally, we demonstrate the power and efficiency of our approach by comparing our algorithms' performance on the link prediction task against an array of current baseline methods across three distinct real-world dynamic networks.