The central premise of this book is that value at the enterprise is created by making decisions, not with data or predictive technologies alone. Nonetheless, we can piggyback on the big data and AI revolutions and start making better choices in a systematic and scalable way, by transforming our companies into modern AI- and data-driven decision-making enterprises. To make better decisions, we first need to ask the right questions, forcing us to move from descriptive & predictive analyses to prescriptive courses of action. I devote the first few chapters to clarifying these concepts and explaining how to ask better business questions suitable for this type of analysis. I then delve into the anatomy of decision-making, starting with the consequences or outcomes we want to achieve, moving backward to the actions we can take, and discussing the problems and opportunities created by intervening uncertainty and causality.

Market Basket Analysis (MBA) is a popular technique to identify associations between products, which is crucial for business decision making. Previous studies typically adopt conventional frequent itemset mining algorithms to perform MBA. However, they generally fail to uncover rarely occurring associations among the products at their most granular level. Also, they have limited ability to capture temporal dynamics in associations between products. Hence, we propose OMBA, a novel representation learning technique for Online Market Basket Analysis. OMBA jointly learns representations for products and users such that they preserve the temporal dynamics of product-to-product and user-to-product associations. Subsequently, OMBA proposes a scalable yet effective online method to generate products' associations using their representations. Our extensive experiments on three real-world datasets show that OMBA outperforms state-of-the-art methods by as much as 21%, while emphasizing rarely occurring strong associations and effectively capturing temporal changes in associations.

Amazon SageMaker is a fully managed service that allows you to build, train, and deploy machine learning (ML) models quickly. Amazon SageMaker removes the heavy lifting from each step of the ML process to make it easier to develop high-quality models. In August 2019, Amazon SageMaker announced the availability of the pre-installed R kernel in all Regions. This capability is available out-of-the-box and comes with the reticulate library pre-installed. This library offers an R interface for the Amazon SageMaker Python SDK, which enables you to invoke Python modules from within an R script.

With rapid advancements in machine learning (ML) techniques over the past decade, intelligent decision-making and prediction systems are poised to transform productivity and lead to significant economic gains. A study conducted by PwC Global concludes that by the end of this decade, the total positive impact of artificial intelligence (AI) on the global economy could be above $15 trillion, driven mostly by enhancements in consumer products. To make that happen, however, businesses must make strategic investments in the type of technology that moves AI projects into production (productionizing) and helps customers deploy them. Unfortunately, PwC's survey reveals the percentage of executives planning to deploy AI has gone down from 20 percent a year ago to only 4 percent at the beginning of 2020. The primary reason for this decrease is the gap between the growing volume of data and data-driven modeling capabilities, and the necessary skills and toolsets.

Determinantal point processes (DPPs) have attracted significant attention from the machine learning community for their ability to model subsets drawn from a large collection of items. Recent work shows that nonsymmetric DPP kernels have significant advantages over symmetric kernels in terms of modeling power and predictive performance. However, the nonsymmetric kernel learning algorithm from prior work has computational complexity that is cubic in the size of the DPP ground set, from which subsets are drawn, making it impractical to use at large scales. In this work, we propose a new decomposition for nonsymmetric DPP kernels that induces linear-time complexity for learning and approximate maximum a posteriori (MAP) inference. We also prove a lower bound on the quality of this MAP approximation. Through evaluation on real-world datasets, we show that our new decomposition not only scales better, but also matches or exceeds the predictive performance of prior work.

AWS DeepRacer is a fun and easy way for developers with no prior experience to get started with machine learning (ML). At the end of the 2019 season, the AWS DeepRacer League engaged the Amazon ML Solutions Lab to develop a new sports analytics feature for the AWS DeepRacer Championship Cup at re:Invent 2019. The purpose for these real-time analytics was to provide context and more in-depth experience with top competitors' strategies and tactics. This helped viewers tangibly interpret how specific model strategy translated to on-track performance, which further demystified ML development and demonstrated its real-world application. This enhancement enabled fans to monitor the performance and driving style of competitors from around the world.

Machine learning (ML) is gaining momentum across a number of industries and scenarios as enterprises look to drive innovation, increase efficiency, and reduce costs. Microsoft Azure Machine Learning empowers developers and data scientists with enterprise-grade capabilities to accelerate the ML lifecycle. At Microsoft Build 2020, we announced several advances to Azure Machine Learning across the following areas: ML for all skills, Enterprise grade MLOps, and responsible ML. New enhancements provide ML access for all skills. Data scientists and developers can now access an enhanced notebook editor directly inside Azure Machine Learning studio.

We study the problem of modeling purchase of multiple items and utilizing it to display optimized recommendations, which is a central problem for online e-commerce platforms. Rich personalized modeling of users and fast computation of optimal products to display given these models can lead to significantly higher revenues and simultaneously enhance the end user experience. We present a parsimonious multi-purchase family of choice models called the BundleMVL-K family, and develop a binary search based iterative strategy that efficiently computes optimized recommendations for this model. This is one of the first attempts at operationalizing multi-purchase class of choice models. We characterize structural properties of the optimal solution, which allow one to decide if a product is part of the optimal assortment in constant time, reducing the size of the instance that needs to be solved computationally. We also establish the hardness of computing optimal recommendation sets. We show one of the first quantitative links between modeling multiple purchase behavior and revenue gains. The efficacy of our modeling and optimization techniques compared to competing solutions is shown using several real world datasets on multiple metrics such as model fitness, expected revenue gains and run-time reductions. The benefit of taking multiple purchases into account is observed to be $6-8\%$ in relative terms for the Ta Feng and UCI shopping datasets when compared to the MNL model for instances with $\sim 1500$ products. Additionally, across $8$ real world datasets, the test log-likelihood fits of our models are on average $17\%$ better in relative terms. The simplicity of our models and the iterative nature of our optimization technique allows practitioners meet stringent computational constraints while increasing their revenues in practical recommendation applications at scale.

Over the last few years, there has been lots of conversation around how AI is going to affect our lives, and the economy. A quick google search returns multiple studies. For example, PWC says that the UK's economy GDP will increase by at least 5% as the result of AI. A report by McKinsey says that the annual productivity will be increasing by 1.2% each year. There are also some books that have started to come out on that subject. I am also talking about this topic in my upcoming book called Uncertainty.

Massachusetts-based supermarket chain Market Basket has announced plans for a second Rhode Island store. The 89,000-square-foot store in Warwick expected to open next year will be located at a site that was previously home to a Sam's Club and later an At Home store, according to a statement from Mayor Joseph Solomon and Market Basket President and CEO Arthur T. Demoulas. "Our city's central location in the state, combined with our growing business climate, continue to make Warwick a natural choice for multiple companies looking to expand their reach in the Ocean State," Solomon said in a statement. Privately-owned Market Basket currently has 81 stores in Massachusetts, New Hampshire and Maine. The company in March announced plans for a store in Johnston.