Benjamin Recht – författare

How the computer revolution shaped our conception of rationality—and why human problems require solutions rooted in human intuition, morality, and judgmentIn the 1940s, mathematicians set out to design computers that could act as ideal rational agents in the face of uncertainty. The Irrational Decision tells the story of how they settled on a peculiar mathematical definition of rationality in which every decision is a statistical question of risk. Benjamin Recht traces how this quantitative standard came to define our understanding of rationality, looking at the history of optimization, game theory, statistical testing, and machine learning. He explains why, now more than ever, we need to resist efforts by powerful tech interests to drive public policy and essentially rule our lives.While mathematical rationality has proven valuable in accelerating computers, regulating pharmaceuticals, and deploying electronic commerce, it fails to solve messy human problems and has given rise to a view of a rational world that is not only overquantified but surprisingly limited. Recht shows how these mathematical methods emerged from wartime research and influenced fields ranging from economics to health care, drawing on illuminating examples ranging from diet planning to chess to self-driving cars.Highlighting both the power and limitations of mathematical rationality, The Irrational Decision reveals why only humans can resolve fundamentally political or value-based questions and proposes a more expansive approach to decision making that is appropriately supported by computational tools yet firmly rooted in human intuition, morality, and judgment.

An authoritative, up-to-date graduate textbook on machine learning that highlights its historical context and societal impactsPatterns, Predictions, and Actions introduces graduate students to the essentials of machine learning while offering invaluable perspective on its history and social implications. Beginning with the foundations of decision making, Moritz Hardt and Benjamin Recht explain how representation, optimization, and generalization are the constituents of supervised learning. They go on to provide self-contained discussions of causality, the practice of causal inference, sequential decision making, and reinforcement learning, equipping readers with the concepts and tools they need to assess the consequences that may arise from acting on statistical decisions.Provides a modern introduction to machine learning, showing how data patterns support predictions and consequential actionsPays special attention to societal impacts and fairness in decision makingTraces the development of machine learning from its origins to todayFeatures a novel chapter on machine learning benchmarks and datasetsInvites readers from all backgrounds, requiring some experience with probability, calculus, and linear algebraAn essential textbook for students and a guide for researchers

Optimization techniques are at the core of data science, including data analysis and machine learning. An understanding of basic optimization techniques and their fundamental properties provides important grounding for students, researchers, and practitioners in these areas. This text covers the fundamentals of optimization algorithms in a compact, self-contained way, focusing on the techniques most relevant to data science. An introductory chapter demonstrates that many standard problems in data science can be formulated as optimization problems. Next, many fundamental methods in optimization are described and analyzed, including: gradient and accelerated gradient methods for unconstrained optimization of smooth (especially convex) functions; the stochastic gradient method, a workhorse algorithm in machine learning; the coordinate descent approach; several key algorithms for constrained optimization problems; algorithms for minimizing nonsmooth functions arising in data science; foundations of the analysis of nonsmooth functions and optimization duality; and the back-propagation approach, relevant to neural networks.

Optimization techniques are at the core of data science, including data analysis and machine learning. An understanding of basic optimization techniques and their fundamental properties provides important grounding for students, researchers, and practitioners in these areas. This text covers the fundamentals of optimization algorithms in a compact, self-contained way, focusing on the techniques most relevant to data science. An introductory chapter demonstrates that many standard problems in data science can be formulated as optimization problems. Next, many fundamental methods in optimization are described and analyzed, including: gradient and accelerated gradient methods for unconstrained optimization of smooth (especially convex) functions; the stochastic gradient method, a workhorse algorithm in machine learning; the coordinate descent approach; several key algorithms for constrained optimization problems; algorithms for minimizing nonsmooth functions arising in data science; foundations of the analysis of nonsmooth functions and optimization duality; and the back-propagation approach, relevant to neural networks.