ZenodoMachine Learning Basics: 2024 Reading Recommendations
The AI bubble craze spared no one with even a peripheral involvement in the arts, humanities, and sciences, and my own field proved no exception, with the spectrum of reactions ranging from genuine excitement at the prospect of leveraging large datasets, to existential anxiety about professional displacement, to outright ideological opposition or uncritical enthusiasm. The reality, as is more often than not the case, settled somewhere within those bounds and proved considerably less dramatic than the surrounding noise suggested.
I initially dismissed both my colleagues' reactions and the broader panic, having taken several courses and worked directly with neural networks myself, which gave me the perhaps unwarranted confidence that I had inoculated myself against the craze. But when the practical necessity of integrating machine learning into our workflow became unavoidable and the barrage of polarized opinions intensified, it became clear that the prior effort had to be directed toward familiarizing colleagues with machine learning as a concrete and tractable tool rather than an abstraction.
In the latter part of 2024 and through the first month of 2025, with funding secured for an internal ML working group within the lab, I curated a reading list of selected chapters and passages intended to circulate among colleagues and begin the iterative work of demystifying the technology, tempering their perspective, and providing a solid foundation from which they could not only reassess their position on AI but also begin thinking practically about which menial and repetitive tasks in our workflow warranted automation with something more adaptive than a static algorithm.
Funding pressures and broader institutional circumstances have since delayed the ML group indefinitely, but I have no intention of letting the book list dissolve quietly, particularly now that 2025 has already produced several releases worth adding to it. I will share the updated version once a few other ongoing projects have cleared and I'm through the ML TBR.
Machine Learning by Ethem Alpaydin
I consider this the optimal entry point for any non-specialist with a STEM background and a basic familiarity with traditional programming, statistics, and mathematics. Alpaydin builds a genuine understanding of machine learning's capacity through real-world examples and well-chosen metaphors, and draws with particular clarity the distinctions between AI, ML, and deep learning. I have a detailed review published with the Fulbright Chronicles journal covering this title.
The Hundred-Page Machine Learning Book by Andriy Burkov
A short, technically rigorous treatment of the algorithms, principles, and concepts that most reliably confound the non-specialist. The language is more demanding and not always immediately accessible, but the book earns its status as a staple, if only for the intellectual honesty with which it delineates what falls outside its own scope. After Alpaydin, this serves as both a refresher and a more granular dive into the field's foundations.
Radically Human: How New Technology Is Transforming Business and Shaping Our Future by Paul R. Daugherty and H. James Wilson
One of the more measured and intellectually grounded business books on AI, built around the premise that humans and AI are fundamentally inseparable and mutually constitutive rather than adversarial. Whether through climate modeling, healthcare outcomes, or pricing efficiency, the argument is that neither functions optimally without the other. The book also proposes the IDEAS framework, standing for Intelligence, Data, Expertise, Architecture, and Strategy, as a human-centered approach to building a coherent digital society. What distinguishes it from similar titles is the breadth of its case studies across industries as varied as electronics, fashion, healthcare, and insurance, each contextualized sufficiently to make the concepts land rather than float.
Artificial Intelligence: A Guide for Thinking Humans by Melanie Mitchell
A grounding read that systematically dismantles the more overwrought scenarios circulating in the collective imagination about AI's capabilities and trajectory. Mitchell is a lifelong researcher and practitioner, and her explanations of the underlying techniques are admirably lucid, accompanied by a notably conservative assessment of what current systems can and cannot do relative to even the most elementary dimensions of human cognition. The one caveat worth stating plainly is that the book was written in 2019 and the field has shifted considerably since, which makes one genuinely curious whether recent developments have revised any of her positions.
AI Superpowers: China, Silicon Valley, and the New World Order by Kai-Fu Lee
An intentional departure from the Western-centric defaults that dominate most AI discourse, and a necessary one. The book is neither technical nor alarmist, but it is genuinely illuminating on the structural and cultural differences between the American and Chinese approaches to AI development, including the instructive contrast between Silicon Valley's light-touch model and China's heavier, more infrastructurally embedded one. Lee's observation that US companies consistently fail in China not because of government interference but because of a stubborn refusal to localize their platforms is worth sitting with. His argument that AI will substantially reshape labor markets, to the point of making universal basic income a serious policy conversation, was not something I had previously connected to the technology's trajectory in such concrete terms. Where the book falls short, and it is a real gap, is in its near-silence on the ethical dimensions of training data: if the data carries bias and the optimization target is purely commercial, the model will scale those biases with neither awareness nor friction, a risk that may materialize well before the macro-level transformations Lee envisions.
Artificial Intelligence: The Basics by Kevin Warwick
A more technical introduction to the field, useful primarily as a consolidating read after the preceding titles, offering a high-level survey of AI's principal methods and subfields alongside recommendations for deeper reading on each. Treat it as a well-organized map rather than a destination in itself.
Machine Learning for Beginners by Chris Sebastian
Included largely because a meaningful contingent of ML practitioners cite it as a formative text, and because if Alpaydin and Burkov did not find purchase with a given reader, this one might. It also brings IoT into the conversation, which the other titles largely leave untouched and which is not an irrelevant dimension of where applied ML is actually headed.
I initially dismissed both my colleagues' reactions and the broader panic, having taken several courses and worked directly with neural networks myself, which gave me the perhaps unwarranted confidence that I had inoculated myself against the craze. But when the practical necessity of integrating machine learning into our workflow became unavoidable and the barrage of polarized opinions intensified, it became clear that the prior effort had to be directed toward familiarizing colleagues with machine learning as a concrete and tractable tool rather than an abstraction.
In the latter part of 2024 and through the first month of 2025, with funding secured for an internal ML working group within the lab, I curated a reading list of selected chapters and passages intended to circulate among colleagues and begin the iterative work of demystifying the technology, tempering their perspective, and providing a solid foundation from which they could not only reassess their position on AI but also begin thinking practically about which menial and repetitive tasks in our workflow warranted automation with something more adaptive than a static algorithm.
Funding pressures and broader institutional circumstances have since delayed the ML group indefinitely, but I have no intention of letting the book list dissolve quietly, particularly now that 2025 has already produced several releases worth adding to it. I will share the updated version once a few other ongoing projects have cleared and I'm through the ML TBR.
Machine Learning by Ethem Alpaydin
I consider this the optimal entry point for any non-specialist with a STEM background and a basic familiarity with traditional programming, statistics, and mathematics. Alpaydin builds a genuine understanding of machine learning's capacity through real-world examples and well-chosen metaphors, and draws with particular clarity the distinctions between AI, ML, and deep learning. I have a detailed review published with the Fulbright Chronicles journal covering this title.
The Hundred-Page Machine Learning Book by Andriy Burkov
A short, technically rigorous treatment of the algorithms, principles, and concepts that most reliably confound the non-specialist. The language is more demanding and not always immediately accessible, but the book earns its status as a staple, if only for the intellectual honesty with which it delineates what falls outside its own scope. After Alpaydin, this serves as both a refresher and a more granular dive into the field's foundations.
Radically Human: How New Technology Is Transforming Business and Shaping Our Future by Paul R. Daugherty and H. James Wilson
One of the more measured and intellectually grounded business books on AI, built around the premise that humans and AI are fundamentally inseparable and mutually constitutive rather than adversarial. Whether through climate modeling, healthcare outcomes, or pricing efficiency, the argument is that neither functions optimally without the other. The book also proposes the IDEAS framework, standing for Intelligence, Data, Expertise, Architecture, and Strategy, as a human-centered approach to building a coherent digital society. What distinguishes it from similar titles is the breadth of its case studies across industries as varied as electronics, fashion, healthcare, and insurance, each contextualized sufficiently to make the concepts land rather than float.
Artificial Intelligence: A Guide for Thinking Humans by Melanie Mitchell
A grounding read that systematically dismantles the more overwrought scenarios circulating in the collective imagination about AI's capabilities and trajectory. Mitchell is a lifelong researcher and practitioner, and her explanations of the underlying techniques are admirably lucid, accompanied by a notably conservative assessment of what current systems can and cannot do relative to even the most elementary dimensions of human cognition. The one caveat worth stating plainly is that the book was written in 2019 and the field has shifted considerably since, which makes one genuinely curious whether recent developments have revised any of her positions.
AI Superpowers: China, Silicon Valley, and the New World Order by Kai-Fu Lee
An intentional departure from the Western-centric defaults that dominate most AI discourse, and a necessary one. The book is neither technical nor alarmist, but it is genuinely illuminating on the structural and cultural differences between the American and Chinese approaches to AI development, including the instructive contrast between Silicon Valley's light-touch model and China's heavier, more infrastructurally embedded one. Lee's observation that US companies consistently fail in China not because of government interference but because of a stubborn refusal to localize their platforms is worth sitting with. His argument that AI will substantially reshape labor markets, to the point of making universal basic income a serious policy conversation, was not something I had previously connected to the technology's trajectory in such concrete terms. Where the book falls short, and it is a real gap, is in its near-silence on the ethical dimensions of training data: if the data carries bias and the optimization target is purely commercial, the model will scale those biases with neither awareness nor friction, a risk that may materialize well before the macro-level transformations Lee envisions.
Artificial Intelligence: The Basics by Kevin Warwick
A more technical introduction to the field, useful primarily as a consolidating read after the preceding titles, offering a high-level survey of AI's principal methods and subfields alongside recommendations for deeper reading on each. Treat it as a well-organized map rather than a destination in itself.
Machine Learning for Beginners by Chris Sebastian
Included largely because a meaningful contingent of ML practitioners cite it as a formative text, and because if Alpaydin and Burkov did not find purchase with a given reader, this one might. It also brings IoT into the conversation, which the other titles largely leave untouched and which is not an irrelevant dimension of where applied ML is actually headed.