There are lots of education options available online, provided you're a self-starter with the discipline to do a lot of coursework on your own. For example, Microsoft's AI School offers a variety of lessons for developers in everything from text analytics and object recognition to custom neural-network models. The content is angled toward data scientists and developers, and heavily emphasizes the use of Microsoft products (of course) in addition to "universal" A.I. skills. It's also free, although those who want Verified Certificates will need to pay a fee. Microsoft, of course, is far from your only option when it comes to learning about A.I. online, particularly with regard to beginner-level material.

MOUNTAIN VIEW, CA – Across every industry, whether it is manufacturing, healthcare, financial services, and more, companies and organizations of all sizes are realizing that AI has the ability to generate amazing value, transform processes, and uncover insights for smarter decisionmaking. But behind the algorithms that are empowering the use cases are the data scientists who drive the models and analyze the insights. According to a report from IBM, Burning Glass and Business Higher Education Forum, the number of job openings for data and analytics talent will increase by 364,000 to 2,720,000 in 2020. Millions of staff will be needed to fuel the ever-growing demand for AI and machine learning. As a result, there's a growing gap between the supply and demand of AI talent, a gap which has resulted in increasingly higher salaries for those in the field.

Data science internships are highly sought after, as studying machine learning in a classroom setting cannot be compared to applying it to solve real client problems. Yet, hopefuls for data science internships will need to have more than just technical skills under their belts, said Mr Ng Sing Kwan, vice president, data science at e-commerce company Lazada. "A lot of people think that to be a good data scientist, they just need to have good coding skills and understand machine learning statistical concepts. But what they lack--and what we look out for--are problem-solving skills and business acumen," he said. Mr Ng, who transited from a business background to data science, was speaking at a panel discussion titled'What Does It Take To Land The First Data Science Job/Internship?', organised by volunteer-led data analytics community Analytics.Club Singapore in partnership with SGInnovate.

At Spark AI Summit in June, we announced MLflow, an open-source platform for the complete machine learning cycle. The platform's philosophy is simple: work with any popular machine learning library; allow machine learning developers experiment with their models, preserve the training environment, parameters, and dependencies, and reproduce their results; and finally deploy, monitor and serve them seamlessly--all in an open manner with limited constraints. In this blog, we will focus on one of the factors: Minimal time to get started. In upcoming blogs, we will elaborate on the other factors, albeit we'll briefly mention them here. Let's consider the level of effort it takes to get started using MLflow in your favorite IDE.

Honor, a leading smartphone e-brand under the Huawei Group signed up to work with Central Saint Martins (CSM) students to explore the concept of colour and its emotional significance within the history of design, as well as the relationship between art, design and technological innovation. Collaborating to find beauty in Artificial Intelligence (AI), selected CSM students are to be awarded the Honor Art prize, which is an award given for a piece of final degree work that celebrates the innovative use of colour and technology in artistic practice. This year, the inaugural prize was awarded to MA Fine Art student Marco Pantaleoni for his series of works in 3D scanning, photography and painting. Receiving his award, Marco said: "I feel honoured to be awarded this prize. Technology is an essential part of my practice, so to be recognised by a technology brand like Honor, not only reinforces some of the concepts behind my work, it also really resonates with the way I create."

YOU'VE heard of it in movies or in passing conversations. Maybe your workplace uses it, or you're considering using it yourself. As technology continues to make ripples across the workplace, AI has become increasingly prevalent. Through AI, companies are able to analyse large amounts of data, which will allow them to better engage with customers. Today, AI is easily accessible.

The ability to learn from incrementally arriving data is essential for any life-long learning system. However, standard deep neural networks forget the knowledge about the old tasks, a phenomenon called catastrophic forgetting, when trained on incrementally arriving data. We discuss the biases in current Generative Adversarial Networks (GAN) based approaches that learn the classifier by knowledge distillation from previously trained classifiers. These biases cause the trained classifier to perform poorly. We propose an approach to remove these biases by distilling knowledge from the classifier of AC-GAN. Experiments on MNIST and CIFAR10 show that this method is comparable to current state of the art rehearsal based approaches. The code for this paper is available at https://bit.ly/incremental-learning

Raven's Progressive Matrices are one of the widely used tests in evaluating the human test taker's fluid intelligence. Analogously, this paper introduces geometric generalization based zero-shot learning tests to measure the rapid learning ability and the internal consistency of deep generative models. Our empirical research analysis on state-of-the-art generative models discern their ability to generalize concepts across classes. In the process, we introduce Infinite World, an evaluable, scalable, multi-modal, light-weight dataset and Zero-Shot Intelligence Metric ZSI. The proposed tests condenses human-level spatial and numerical reasoning tasks to its simplistic geometric forms. The dataset is scalable to a theoretical limit of infinity, in numerical features of the generated geometric figures, image size and in quantity. We systematically analyze state-of-the-art model's internal consistency, identify their bottlenecks and propose a pro-active optimization method for few-shot and zero-shot learning.

We present an algorithm to generate synthetic datasets of tunable difficulty on classification of Morse code symbols for supervised machine learning problems, in particular, neural networks. The datasets are spatially one-dimensional and have a small number of input features, leading to high density of input information content. This makes them particularly challenging when implementing network complexity reduction methods. We explore how network performance is affected by deliberately adding various forms of noise and expanding the feature set and dataset size. Finally, we establish several metrics to indicate the difficulty of a dataset, and evaluate their merits. The algorithm and datasets are open-source.

Due to the iterative nature of most nonnegative matrix factorization (\textsc{NMF}) algorithms, initialization is a key aspect as it significantly influences both the convergence and the final solution obtained. Many initialization schemes have been proposed for NMF, among which one of the most popular class of methods are based on the singular value decomposition (SVD). However, these SVD-based initializations do not satisfy a rather natural condition, namely that the error should decrease as the rank of factorization increases. In this paper, we propose a novel SVD-based \textsc{NMF} initialization to specifically address this shortcoming by taking into account the SVD factors that were discarded to obtain a nonnegative initialization. This method, referred to as nonnegative SVD with low-rank correction (NNSVD-LRC), allows us to significantly reduce the initial error at a negligible additional computational cost using the low-rank structure of the discarded SVD factors. NNSVD-LRC has two other advantages compared to previous SVD-based initializations: (1) it provably generates sparse initial factors, and (2) it is faster as it only requires to compute a truncated SVD of rank $\lceil r/2 + 1 \rceil$ where $r$ is the factorization rank of the sought NMF decomposition (as opposed to a rank-$r$ truncated SVD for other methods). We show on several standard dense and sparse data sets that our new method competes favorably with state-of-the-art SVD-based initializations for NMF.