Comment Policy: Please write your comments that match the topic of this page's posts. Comments that contain links will not be displayed until they are approved.

Alice decides to do some quick analysis on the trends using Kaggle Data Science survey to see what backgrounds do the current Data Science practitioners have. A majority of data scientists have college degrees, infact a majority of them have a Masters degree. So Alice would do well to go to college. But Alice is also curious of the importance of getting a degree if she wants her dream job in her dream country. Let's look at those patterns.

Parameter space exploration methods with black-box optimization have recently been shown to outperform state-of-the-art approaches in continuous control reinforcement learning domains. In this paper, we examine reasons why these methods work better and the situations in which they are worse than traditional action space exploration methods. Through a simple theoretical analysis, we show that when the parametric complexity required to solve the reinforcement learning problem is greater than the product of action space dimensionality and horizon length, exploration in action space is preferred. This is also shown empirically by comparing simple exploration methods on several toy problems.

We introduce manifold-modeling flows (MFMFs), a new class of generative models that simultaneously learn the data manifold as well as a tractable probability density on that manifold. Combining aspects of normalizing flows, GANs, autoencoders, and energy-based models, they have the potential to represent data sets with a manifold structure more faithfully and provide handles on dimensionality reduction, denoising, and out-of-distribution detection. We argue why such models should not be trained by maximum likelihood alone and present a new training algorithm that separates manifold and density updates. With two pedagogical examples we demonstrate how manifold-modeling flows let us learn the data manifold and allow for better inference than standard flows in the ambient data space.

We tackle some fundamental problems in probability theory on corrupted random processes on the integer line. We analyze when a biased random walk is expected to reach its bottommost point and when intervals of integer points can be detected under a natural model of noise. We apply these results to problems in learning thresholds and intervals under a new model for learning under adversarial design.

Deep-learning-as-a-service is a novel and promising computing paradigm aiming at providing machine/deep learning solutions and mechanisms through Cloud-based computing infrastructures. Thanks to its ability to remotely execute and train deep learning models (that typically require high computational loads and memory occupation), such an approach guarantees high performance, scalability, and availability. Unfortunately, such an approach requires to send information to be processed (e.g., signals, images, positions, sounds, videos) to the Cloud, hence having potentially catastrophic-impacts on the privacy of users. This paper introduces a novel distributed architecture for deep-learning-as-a-service that is able to preserve the user sensitive data while providing Cloud-based machine and deep learning services. The proposed architecture, which relies on Homomorphic Encryption that is able to perform operations on encrypted data, has been tailored for Convolutional Neural Networks (CNNs) in the domain of image analysis and implemented through a client-server REST-based approach. Experimental results show the effectiveness of the proposed architecture.

Knowledge distillation (KD), i.e. one classifier being trained on the outputs of another classifier, is an empirically very successful technique for knowledge transfer between classifiers. It has even been observed that classifiers learn much faster and more reliably if trained with the outputs of another classifier as soft labels, instead of from ground truth data. However, there has been little or no theoretical analysis of this phenomenon. We provide the first theoretical analysis of KD in the setting of extremely wide two layer non-linear networks in model and regime in (Arora et al., 2019; Du & Hu, 2019; Cao & Gu, 2019). We prove results on what the student network learns and on the rate of convergence for the student network. Intriguingly, we also confirm the lottery ticket hypothesis (Frankle & Carbin, 2019) in this model. To prove our results, we extend the repertoire of techniques from linear systems dynamics. We give corresponding experimental analysis that validates the theoretical results and yields additional insights.

Enver Yucel is the founder of BAU Global, a broad education network headquartered in Turkey, consisting of five universities, three language schools, four academic centers and one boarding school spread across North America, Europe, Africa and Asia. Yucel has devoted his life to education, having served an estimated 150,000 students since starting his first institution with three class rooms in Istanbul in 1974. He is also a member of the Advisory Board of the UN Institute for Training and Research. He was invited to speak at the AI World Conference & Expo in Boston in the fall of 2019, the first Turkish speaker in the four years of the conference. He recently took some time to answer questions posed by AI Trends Editor John P Desmond, who was in the audience for his Boston talk.

Free Course - Learn Alexa the fun way [2020] Learn to create interesting and fun-filled Amazon Alexa skills and publish them, Learn Alexa the fun way. Learn to create interesting and fun-filled Amazon Alexa skills and publish them. Learn Alexa the fun way About this Course Hello learners, Welcome to MAKERDEMY's "Learn Alexa the fun way" course. If you are looking for that one course that will help you gain the confidence to build and publish Amazon Alexa skills, you have come to the right place. With numerous custom made illustrations and animations, we have set the standard in terms of production quality.

AI4ALL Open Learning is a free program which community-based organisations, as well as teachers, can use the resource to educate the community and high-school students about Artificial Intelligence. Recently, AI4ALL announced a new module which high school teachers with a free online lesson to engage students around AI's role in the current crisis. In the program, the ExploreAI curriculum can be implemented in 10, 20 or 30 hrs. The reason behind this program is to support the community by sharing relevant, free curriculum and teaching resources. AI4All is trying every possible way to ensure the best to serve all people in the work to increase diversity and inclusion in artificial intelligence.