Artificial Intelligence will define the next generation of software solutions. This computer science course provides an overview of AI, and explains how it can be used to build smart apps that help organizations be more efficient and enrich people's lives. It uses a mix of engaging lectures and hands-on activities to help you take your first steps in the exciting field of AI. Discover how machine learning can be used to build predictive models for AI. Learn how software can be used to process, analyze, and extract meaning from natural language; and to process images and video to understand the world the way we do.

Security has risen to the top of the priority list for many managers in institutional and commercial facilities. Nowhere is the issue more important that in K-12 schools, given the growing threat that mass shooters present to students, staff and faculty. School district officials have revisited every conceivable angle of the issue as it relates to facilities, from building design to access control. Now some managers are taking a closer look at emerging technology to give them an edge. Artificial intelligence (AI) is helping to change security cameras in school districts from passive to active observers, according to VOA News. They can identify people, suspicious behavior and guns and gather large amounts of information.

One Georgia school district plans to spend $16.5 million to install artificial intelligence-powered surveillance cameras in its roughly 100 buildings in coming years. Businesses across almost every industry deploy artificial intelligence to make jobs simpler for staff and tasks easier for consumers. Computer software teaches customer service agents how to be more compassionate, schools use machine learning to scan for weapons and mass shooters on campus, and doctors use AI to map the root cause of diseases. Sectors such as cybersecurity, online entertainment and retail use the tech in combination with wide swaths of customer data in revolutionary ways to streamline services. Though these applications may seem harmless, perhaps even helpful, the AI is only as good as the information fed into it, which can have serious implications.

Predictive modeling, or supervised machine learning, is a powerful tool for using data to make predictions about the world around us. Once you understand the basic ideas of supervised machine learning, the next step is to practice your skills so you know how to apply these techniques wisely and appropriately. In this course, you will work through four case studies using data from the real world; you will gain experience in exploratory data analysis, preparing data so it is ready for predictive modeling, training supervised machine learning models, and evaluating those models. To take this course, you need some familiarity with tidyverse packages like dplyr and ggplot2 and exposure to machine learning basics.

I'm beyond excited to be here in London with my Microsoft colleagues, as well as innovators, researchers, experts and business decision-makers from around the world at Future Decoded. Over the next two days, we will hear inspiring stories about the possibilities that exist for artificial intelligence to transform the future of work in every industry – and how critical it is that businesses foster a culture that includes everyone as we search for ways to incorporate AI responsibly. This morning's announcement that Microsoft is collaborating with Novartis to use AI to develop treatments and medications faster has the potential to improve patients' lives across the globe. A critical component of our work together is the commitment by Novartis to take AI across the entire organization. This will enable Novartis to bring together previously siloed data sets and research, and to use AI to build upon existing work quickly and efficiently.

Data Science has gained a lot of popularity in the last few years. This field's primary focus is to convert meaningful data into marketing and business strategies which helps a company grow. The data is stored and researched to get in a logical solution. Previously only the top IT companies were involved in this field but today businesses from various sector and fields such as e-commerce, health care, finance, and others are using data analytics. There are various tools available for data analytics such as Hadoop, R programming, SAS, SQL and many more. However the most popular and easy to use tools for data analytics is Python.

Artificial Intelligence (alternate unit) was written and developed by Beverly Clarke. She is author of the book "Computer Science Teacher – insight into the computing classroom." Additionally, she is an Education consultant and former teacher. In writing this unit the following are acknowledged for their contributions in proof reading, checking for technical accuracy, testing activities in the classroom, filming, being sound boards and committed to seeing an AI curriculum available for high school students – Mike Mendelson (NVIDIA), James McClung (formerly of NVIDIA), Joanna Goode (University of Oregon), Alison Lowndes (NVIDIA), Rosie Lane (South Wilts Grammar School for Girls), Peter McOwan (Queen Mary University of London), Paul Curzon (Queen Mary University of London), Liz Austin (NVIDIA), Gemma Bond (Screen Boo Productions) and Neil Rickus (University of Hertfordshire). Morals and Ethics supporting cards were sampled from material by Andrew Csizmadia (Newman University).

Campuses are using artificial intelligence technology to respond and advise, on-demand, to student, faculty, and staff via texting or auditory. These tools extend the accessibility of student support services 24/7 and engage students in a manner most conducive to them.

Wide adoption of complex RNN based models is hindered by their inference performance, cost and memory requirements. To address this issue, we develop AntMan, combining structured sparsity with low-rank decomposition synergistically, to reduce model computation, size and execution time of RNNs while attaining desired accuracy. AntMan extends knowledge distillation based training to learn the compressed models efficiently. Our evaluation shows that AntMan offers up to 100x computation reduction with less than 1pt accuracy drop for language and machine reading comprehension models. Our evaluation also shows that for a given accuracy target, AntMan produces 5x smaller models than the state-of-art. Lastly, we show that AntMan offers super-linear speed gains compared to theoretical speedup, demonstrating its practical value on commodity hardware.

Distillation is a method to transfer knowledge from one model to another and often achieves higher accuracy with the same capacity. In this paper, we aim to provide a theoretical understanding on what mainly helps with the distillation. Our answer is "early stopping". Assuming that the teacher network is overparameterized, we argue that the teacher network is essentially harvesting dark knowledge from the data via early stopping. This can be justified by a new concept, {Anisotropic Information Retrieval (AIR)}, which means that the neural network tends to fit the informative information first and the non-informative information (including noise) later. Motivated by the recent development on theoretically analyzing overparameterized neural networks, we can characterize AIR by the eigenspace of the Neural Tangent Kernel(NTK). AIR facilities a new understanding of distillation. With that, we further utilize distillation to refine noisy labels. We propose a self-distillation algorithm to sequentially distill knowledge from the network in the previous training epoch to avoid memorizing the wrong labels. We also demonstrate, both theoretically and empirically, that self-distillation can benefit from more than just early stopping. Theoretically, we prove convergence of the proposed algorithm to the ground truth labels for randomly initialized overparameterized neural networks in terms of $\ell_2$ distance, while the previous result was on convergence in $0$-$1$ loss. The theoretical result ensures the learned neural network enjoy a margin on the training data which leads to better generalization. Empirically, we achieve better testing accuracy and entirely avoid early stopping which makes the algorithm more user-friendly.