Use code MEETUP at check out for 20% OFF! Spots are limited! Since its establishment in 2009, RMDS has become one of the largest data science communities in California with over 33,000 data professionals and researchers. After the success of over 50 meetups, we have witnessed the growing need for a larger conference with more than 1500 attendees expected. We are glad to announce that RMDS will collaborate with the City of Pasadena, CA to hold its annual conference in Pasadena Convention Center on Dec 6-7. Also, we will launch our unique certificate workshop program with UCR -- one of the most prominent public universities in California -- to provide a practical and hands-on learning experience on Dec 5. Spots are limited.

Multiplicative stochasticity such as Dropout improves the robustness and generalizability of deep neural networks. Here, we further demonstrate that always-on multiplicative stochasticity combined with simple threshold neurons are sufficient operations for deep neural networks. We call such models Neural Sampling Machines (NSM). We find that the probability of activation of the NSM exhibits a self-normalizing property that mirrors Weight Normalization, a previously studied mechanism that fulfills many of the features of Batch Normalization in an online fashion. The normalization of activities during training speeds up convergence by preventing internal covariate shift caused by changes in the input distribution. The always-on stochasticity of the NSM confers the following advantages: the network is identical in the inference and learning phases, making the NSM suitable for online learning, it can exploit stochasticity inherent to a physical substrate such as analog non-volatile memories for in-memory computing, and it is suitable for Monte Carlo sampling, while requiring almost exclusively addition and comparison operations. We demonstrate NSMs on standard classification benchmarks (MNIST and CIFAR) and event-based classification benchmarks (N-MNIST and DVS Gestures). Our results show that NSMs perform comparably or better than conventional artificial neural networks with the same architecture.

Linking authors of short-text contents has important usages in many applications, including Named Entity Recognition (NER) and human community detection. However, certain challenges lie ahead. Firstly, the input short-text contents are noisy, ambiguous, and do not follow the grammatical rules. Secondly, traditional text mining methods fail to effectively extract concepts through words and phrases. Thirdly, the textual contents are temporally skewed, which can affect the semantic understanding by multiple time facets. Finally, using the complementary knowledge-bases makes the results biased to the content of the external database and deviates the understanding and interpretation away from the real nature of the given short text corpus. To overcome these challenges, we devise a neural network-based temporal-textual framework that generates the tightly connected author subgraphs from microblog short-text contents. Our approach, on the one hand, computes the relevance score (edge weight) between the authors through considering a portmanteau of contents and concepts, and on the other hand, employs a stack-wise graph cutting algorithm to extract the communities of the related authors. Experimental results show that compared to other knowledge-centered competitors, our multi-aspect vector space model can achieve a higher performance in linking short-text authors. Additionally, given the author linking task, the more comprehensive the dataset is, the higher the significance of the extracted concepts will be.

Continual lifelong learning is essential to many applications. In this paper, we propose a simple but effective approach to continual deep learning. Our approach leverages the principles of deep model compression, critical weights selection, and progressive networks expansion. By enforcing their integration in an iterative manner, we introduce an incremental learning method that is scalable to the number of sequential tasks in a continual learning process. Our approach is easy to implement and owns several favorable characteristics. First, it can avoid forgetting (i.e., learn new tasks while remembering all previous tasks). Second, it allows model expansion but can maintain the model compactness when handling sequential tasks. Besides, through our compaction and selection/expansion mechanism, we show that the knowledge accumulated through learning previous tasks is helpful to build a better model for the new tasks compared to training the models independently with tasks. Experimental results show that our approach can incrementally learn a deep model tackling multiple tasks without forgetting, while the model compactness is maintained with the performance more satisfiable than individual task training.

Universities are embracing artificial intelligence solutions to assist in IT projects and academics. At George Washington University, after piloting its 24/7 chatbot service MARTHA, 89 percent of users advocated the tool be a permanent tool. "We've created a service broker that can handle decisions on where to go to look for information," Jonathan Fozard, assistant vice president for the CIO's office at George Washington University told EdTech. "As we educated it and users tested it, the Watson component was learning alongside of us. If someone types in a question about 3D printing, we know that's most likely a student who has access to 3D printers in the engineering classroom or a medical enterprise."

Not every remix is an innovation: A network perspective on the 3d-printing community. Is the Maker Movement Contributing to Sustainability? Makers' ambitions to do socially valuable things. An empirically informed taxonomy for the Maker movement. DOI: 10.1007/978-3-319-41267-2_35 Misuraca G., Kucsera, C., Lipparini F., Voigt C., and Radescu R., (2015) ''Mapping and Analysis of ICT-enabled Social Innovation initiatives promoting social investment in integrated approaches to the provision of social services, European Commission's Joint Research Centre, IPTS (Technical Reports).

Nesta Challenges' Longitude Explorer Prize (LEP) engages young people across the UK to think about innovative solutions to society's biggest issues. This year, we're inspiring 11 - 16 year olds to develop an understanding of Artificial Intelligence (A.I.) & how it's impacting the World. Pop along to the Swansea Science Festival - for the chance to find out how to get involved in the Longitude Explorer Prize - the Nesta team will be on hand to chat over interactive fun activities and refreshments.

Have you ever felt overwhelmed by the sheer number of images or documents, or hours of video footage you needed to sort through for a report? Training a machine to do the work for you may be the answer. Learn how artificial intelligence can improve your reporting with the new course from the Knight Center for Journalism in the Americas and instructor John Keefe, "Hands-on Machine Learning Solutions for Journalists." The four-week Big Online Course (BOC) runs from Nov. 18 to Dec. 15, 2019 and costs $95, which includes a certificate for those who successfully complete the course requirements. "At the end of this class, students will have a much better understanding of machine learning. They will actually be able to sort documents, especially images, based on the criteria they set up," said Keefe, who uses these techniques in his work as investigations editor at Quartz.

When I was in seventh grade, we had to take a class called home ec. Everyone brushed it off as a super easy class. "All you have to do is cook and sew," everyone said. One of our first projects, after learning how sewing machines work, was sewing a pair of pajama pants. You'd think it's a pretty simple process.

Machine learning problems can generally be divided into three types.