--Federated learning (FL) is an emerging technique for training machine learning models using geographically dispersed data collected by local entities. It includes local computation and synchronization steps. T o reduce the communication overhead and improve the overall efficiency of FL, gradient sparsification (GS) can be applied, where instead of the full gradient, only a small subset of important elements of the gradient is communicated. Existing work on GS uses a fixed degree of gradient sparsity for i.i.d.-distributed data within a datacenter . In this paper, we consider adaptive degree of sparsity and non-i.i.d. We first present a fairness-aware GS method which ensures that different clients provide a similar amount of updates. Then, with the goal of minimizing the overall training time, we propose a novel online learning formulation and algorithm for automatically determining the near-optimal communication and computation tradeoff that is controlled by the degree of gradient sparsity. The online learning algorithm uses an estimated sign of the derivative of the objective function, which gives a regret bound that is asymptotically equal to the case where exact derivative is available. Experiments with real datasets confirm the benefits of our proposed approaches, showing up to 40% improvement in model accuracy for a finite training time. Modern consumer and enterprise users generate a large amount of data at the network edge, such as sensor measurements from Internet of Things (IoT) devices, images captured by cameras, transaction records of different branches of a company, etc. Such data may not be shareable with a central cloud, due to data privacy regulations and communication bandwidth limitation [1]. In these scenarios, federated learning (FL) is a useful approach for training machine learning models from local data [1]-[5]. The basic process of FL includes local gradient computation at clients and model weight (parameter) aggregation through a server. Instead of sharing the raw data, only model weights or gradients need to be shared between the clients and the server in the FL process.

Efficient human resource management needs accurate assessment and representation of available competences as well as effective mapping of required competences for specific jobs and positions. In this regard, appropriate definition and identification of competence gaps express differences between acquired and required competences. Using a detailed quantification scheme together with a mathematical approach is a way to support accurate competence analytics, which can be applied in a wide variety of sectors and fields. This article describes the combined use of software technologies and mathematical and statistical methods for assessing and analyzing competences in human resource information systems. Based on a standard competence model, which is called a Professional, Innovative and Social competence tree, the proposed framework offers flexible tools to experts in real enterprise environments, either for evaluation of employees towards an optimal job assignment and vocational training or for recruitment processes. The system has been tested with real human resource data sets in the frame of the European project called ComProFITS.

We're bringing TC Sessions: Robotics AI back to UC Berkeley on March 3, and we're excited to announce our jam-packed agenda. For months we've been recruiting speakers from the ranks of the most innovative founders, top technologists and hard-charging VCs working in robotics and AI, and the speaker line-up will capture the remarkable acceleration across the field in the past year. New for this year, we will be hosting our very first pitch-off competition for early-stage robotics companies. There is still time to submit your application. What better way to mark the occasion than to grab an early-bird ticket ($150 savings) right now and right here before prices increase.

"The light bulb was not created by continuously improving the candle." As artificial intelligence and machine learning sweep the global economy, we find innovations from the last century becoming increasingly obsolete. In fact, the world is changing so rapidly that almost every facet of human life has been disrupted -- some more than others. Technology has revolutionized the way we communicate, undertake research, learn, interact with other people, work, travel, access healthcare, and enjoy leisurely activities. According to a report published by Tech Nation,[1] the US is the global leader in technology investments, accounting for 49% (or $149 billion) of the capital raised by tech scale-ups over the last four years (Chinese scale-ups raised 20%).

Indian multinational corporation Wipro, in collaboration with the National Association of Software and Services Companies (Nasscom), will be setting up a platform to train students in emerging technologies -- artificial intelligence (AI), data science, internet of things (IoT) and cyber securities. Wipro has decided to take up the initiative under its corporate social responsibility (CSR) programme, TalentNext, that aims to enhance the quality of engineering education in India by training college faculties and academic leaders, who will later train students. The platform, set up by Wipro and Nasscom, will be looking to train 10K students from over 20 engineering colleges in India. With this platform, the duo wants to build a talent pool of students certified by Wipro and Nasscom and provide them with greater job opportunities. Wipro's Chairman Premji, in a press release, said that the platform will bring together content and people, alongside the focus on curation and learning at one's own pace.

He nailed it in his high-school tests, and nailed it again in his university entrance examination. But, according to headlines, "he wants to be a playwright, not a scientist." Oh, how meaningful a contraposition can be! His was a wonderful achievement, but the bodies of the news stories were not dedicated to how he had been able to accomplish it, what his study techniques are or his motivations. No, they were primarily focused on what he wants and doesn't want to be.

The European Tech Startup's market is booming, thus I decided to remark 10 European Tech Startups to Watch in 2020. You can find their category and the representative country. You also can find them in our Worldwide directory of Tech Startups. In general, learning management systems are cluttered with stiff, confined and hard to scale solution, leaving the learning departments in pain to provide an engaging learning experience to their workforce. Knowledge is now everywhere and most of the time, happens outside of the company.

When a New Year begins, some people write down goals and wishes for the year. Understandably, these are things they wish to accomplish before the year culminates. Oftentimes, there is an overwhelming hype to make resolutions for the year. Personally, I have not been making resolutions but I have always desired to accomplish some targets within the year. In this article, I look back at the year 2019 with excitement as I celebrate the milestones I have made in Data Science.

Students who have at least high school knowledge in math and who want to start learning Machine Learning. Any intermediate level people who know the basics of machine learning, including the classical algorithms like linear regression or logistic regression, but who want to learn more about it and explore all the different fields of Machine Learning. Any people who are not that comfortable with coding but who are interested in Machine Learning and want to apply it easily on datasets. Anyone willing to learn machine learning on Google cloud platform. Any students in college who want to start a career in Data Science. Any data analysts who want to level up in Machine Learning.

In December 2019, I began work on my eBook and this article on eLearning trends in 2020. This is the ninth in a series of articles and eBooks on eLearning trends since 2017. Here is my list of eLearning trends in 2020. I believe all of these will be an integral part of workplace learning in the near future. Given the large number of trends, I have logically grouped them into 3 sections.