The Infinity Machine: Demis Hassabis, DeepMind, and the Quest for Superintelligence
by Sebastian Mallaby; · 30 Mar 2026 · 607pp · 161,998 words
that they could be fed into a computer. Ever since the Dartmouth workshop, artificial intelligence pioneers had wrestled with this conundrum, which philosophers termed the “problem of induction.” But however hard they tried, humans’ mental shortcuts could neither be defined nor written into a program. “AI has utterly failed, over a quarter century
…
game. It would play at superhuman level. Of course, trying out every possible permutation would have taken decades. To shortcut the challenge—to solve the problem of induction—Mnih added in some more deep learning. As the agent collected experiences, each one consisting of the state of the game, the action taken, and
…
or human data. The old obstacle to AI—the impossibility of devising a deductive system to classify and explain the world—had been bypassed. The problem of induction—the challenge of finding patterns in an infinity of data—had been vanquished. “We’re no longer constrained to systems with predefined rules,” Silver observed
…
’s only been eight years.” This was science at digital speed, I suggested, echoing a phrase from Hassabis’s Nobel lecture. The solving of the problem of induction—the invention of machines that could induce patterns in an infinity of data—changed the pace at which science proceeded. AlphaFold heralded infinite discovery, courtesy
…
classical computers to transcend the constraint of binary information. As Turing had foretold, a Turing machine of infinite size could discover infinite patterns, solving the problem of induction and disproving Penrose’s claims about the limits of classical computers. Where Penrose had been fascinated by qubits in fuzzy superposition, because of the multifarious
Prophecy: Prediction, Power, and the Fight for the Future, from Ancient Oracles to AI
by Carissa Véliz · 21 Apr 2026 · 503pp · 129,255 words
. The best we can aim for is a probable argument to try to establish other probable arguments, thereby falling into circularity. Hume’s conclusion—the problem of induction—is that our tendency to project past regularities into the future is not supported by reason. We are not justified in assuming that the future
The Knowledge Machine: How Irrationality Created Modern Science
by Michael Strevens · 12 Oct 2020
of Berlin, and expelled many of its Jewish professors and staff, Reichenbach included. “Then,” Reichenbach is said to have observed, “I understood at last the problem of induction.” REICHENBACH, POPPER, AND many like-minded refugees fleeing the mayhem and malevolence of Central Europe between the wars promoted an ideal of the scientist as
…
now regarded by many as the “father of neuroendocrinology” for his theory of hormone-driven communication within the brain. 22 “I understood at last the problem of induction”: The story is told by Ronald Giere in his book Science without Laws. He attributes it to the philosopher Andreas Kamlah, who, he reports, “knew
The Myth of Artificial Intelligence: Why Computers Can't Think the Way We Do
by Erik J. Larson · 5 Apr 2021
past). The eighteenth-century philosopher David Hume, who first pointed out the limits of induction, gave philosophers and scientists what is now known as the problem of induction. As Hume put it, relying on induction requires us to believe that “instances of which we have had no experience resemble those of which we
…
the world has certain characteristics, and we can examine the world and tease out the knowledge that (we think) we have about it.4 The problem of induction may seem like an armchair worry that philosophers like to indulge, but in fact the limits of inductive inference raise constant problems for scientists in
…
clear inferential rules. To model inference on science, then, we must expose errors in our thinking about scientific investigation and truth-seeking generally. Thus the problem of induction came under his scrutiny; he called it one of the core “problems of philosophy” (in his book of that title), and argued, like Sir Karl
…
floor at 3:30 AM. Of course there are reliable generalizations—we see them everywhere, and it’s not delusional to do so—but the problem of induction, as Russell pointed out, is that we have no grounds for inferring knowledge based only on such generalizations. Science must rely on deeper and more
…
, which is the entire point of the quest to achieve general intelligence. Computer scientists relying on inductive methods often dismiss Hume’s (or Russell’s) problem of induction as irrelevant. As the logic goes, of course there are no guarantees of correctness using induction, but we can get “close enough.” This response misses
…
games, which have rules to box in statistical inferences. A probably approximately correct solution leaves unchanged the problem of induction in dynamic environments outside a game world or research laboratory. AI researchers are aware of the problem of induction (either explicitly or implicitly), but it rarely enters into critiques of machine learning (or deep learning) because
…
. This is like looking for your keys under a lamppost because the light is better there. It’s true that human beings have “solved” the problem of induction well enough to use experience effectively in the real world (where else?). But humans solve the problem of inference not with inductive inference in some
…
knowledge and inference in experience exclusively, which is precisely what machine learning approaches to AI do. We should not be surprised, then, that all the problems of induction bedevil machine learning and data-centric approaches to AI. Data are just observed facts, stored in computers for accessibility. And observed facts, no matter how
…
thought to empower AI systems with previously unavailable “smarts” and insight. In a sense, this is true, but not in the sense necessary to escape problems of induction. We turn to big data next. THE END OF BIG DATA Big data is a notoriously amorphous idea that refers generally to the power of
…
say that data alone, big data or not, and inductive methods like machine learning have inherent limitations that constitute roadblocks to progress in AI. The problem of induction, it turns out, really is a problem for modern AI. Their window into meaning is tied directly to data, which is a limiting constraint on
The Power Makers
by Maury Klein · 26 May 2008 · 782pp · 245,875 words
do all their testing on Sundays because construction work occupied the factory the rest of the week. Long, animated discussions about the potential and technical problems of induction motors, rotary converters, and alternating current itself punctuated their efforts.39 The pressure to produce excited and energized the engineers. Working on the frontier of
Artificial Intelligence: A Modern Approach
by Stuart Russell and Peter Norvig · 14 Jul 2019 · 2,466pp · 668,761 words
Aristotle in 350 BCE in Physics book I, chapter VI: “For the more limited, if adequate, is always preferable.” David Hume (1711–1776) formulated the problem of induction, recognizing that generalizing from examples admits the possibility of errors, in a way that logical deduction does not. He saw that there was no way
…
1 and 3 and h1 is next to a second helix.” 20.5.1An example Recall from Equation (20.5) that the general knowledge-based induction problem is to “solve” the entailment constraint for the unknown Hypothesis, given the Background knowledge and examples described by Descriptions and Classifications. To illustrate this, we
…
, 763 constructive induction algorithms, 760 description, 758 family tree, 760 inverse entailment, 765 inverse resolution, 763, 764–766 Journal of Molecular Biology, 766 knowledge-based induction problem, 758 linear resolution, 765 LINUS system, 765 molecular biology experiments, 767 NEW-LITERALS, 762–763 positive and negative examples, 758, 759 PROGOL system, 765, 766
…
, 763 constructive induction algorithms, 760 description, 758 family tree, 760 inverse entailment, 765 inverse resolution, 763, 764–766 Journal of Molecular Biology, 766 knowledge-based induction problem, 758 linear resolution, 765 LINUS system, 765 molecular biology experiments, 767 NEW-LITERALS, 762–763 positive and negative examples, 758, 759 PROGOL system, 765, 766
Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets
by Nassim Nicholas Taleb · 1 Jan 2001 · 111pp · 1 words
THE RARE-EVENT FALLACY The Mother of All Deceptions Why Don’t Statisticians Detect Rare Events? A Mischievous Child Replaces the Black Balls Seven THE PROBLEM OF INDUCTION FROM BACON TO HUME Cygnus Atratus Niederhoffer SIR KARL’S PROMOTING AGENT Location, Location Popper’s Answer Open Society Nobody Is Perfect Induction and Memory
…
finance professors as they tend to firmly believe that they know something, and something useful at that). It is presented as flowing from Hume’s Problem of Induction (or Aristotle’s inference to the general) as opposed to the paradigm of the gambling literature. In this book probability is principally a branch of
…
evolution are concepts that are misundderstood in the nonbiological world. Life is not continuous. How evolution will be fooled by randomness. A prolegomenon for the problem of induction. SIX: SKEWNESS AND ASYMMETRY We introduce the concept of skewness: Why the terms “bull” and “bear” have limited meaning outside of zoology. A vicious child
…
wrecks the structure of randomness. An introduction to the problem of epistemic opacity. The penultimate step before the problem of induction. SEVEN: THE PROBLEM OF INDUCTION On the chromodynamics of swans. Taking Solon’s warning into some philosophical territory. How Victor Niederhoffer taught me empiricism; I added deduction. Why it
…
more resistant to randomness. Solon also had the intuition of a problem that has obsessed science for the past three centuries. It is called the problem of induction. I call it in this book the black swan or the rare event. Solon even understood another linked problem, which I call the skewness issue
…
of security. The point is dubbed in this book the black swan problem, which we cover in Chapter 7, as it is linked to the problem of induction, a problem that has kept a few thinkers awake at night. It is also related to a problem called denigration of history, as gamblers, investors
…
evolution are concepts that are misunderstood in the nonbiological world. Life is not continuous. How evolution will be fooled by randomness. A prolegomenon for the problem of induction. CARLOS THE EMERGING-MARKETS WIZARD I used to meet Carlos at a variety of New York parties, where he would show up impeccably dressed, though
…
meaning outside of zoology. A vicious child wrecks the structure of randomness. An introduction to the problem of epistemic opacity. The penultimate step before the problem of induction. THE MEDIAN IS NOT THE MESSAGE The essayist and scientist Steven Jay Gould (who, for a while, was my role model), was once diagnosed when
…
entire concept seem like a costly (perhaps very costly) mistake. This leads us to a more fundamental question: The problem of induction, to which we will turn in the next chapter. Seven • THE PROBLEM OF INDUCTION On the chromodynamics of swans. Taking Solon’s warning into some philosophical territory. How Victor Niederhoffer taught me empiricism; I
…
we discuss this problem viewed from the broader standpoint of the philosophy of scientific knowledge. There is a problem in inference well-known as the problem of induction. It is a problem that has been haunting science for a long time, but hard science has not been as harmed by it as the
…
social sciences, particularly economics, even more the branch of financial economics. Why? Because the randomness content compounds its effects. Nowhere is the problem of induction more relevant than in the world of trading—and nowhere has it been as ignored! Cygnus Atratus In his Treatise on Human Nature, the Scots
…
, it was how they knew it, that was the subject of my annoyance. Popper’s Answer Popper came up with a major answer to the problem of induction (to me he came up with the answer). No man has influenced the way scientists do science more than Sir Karl—in spite of the
…
is the reduction in the degree of detected randomness. Pascal’s Wager I conclude with the exposition of my own method of dealing with the problem of induction. The philosopher Pascal proclaimed that the optimal strategy for humans is to believe in the existence of God. For if God exists, then the believer
…
the pot). It would certainly modify the conclusion. It’s a Bull Market As to the second, more serious flaw, I have already discussed the problem of induction. The story focuses on an unusual episode in history; buying its thesis implies accepting that the current returns in asset values are permanent (the sort
…
even know how many colors there are). But somehow people “measure” risks, particularly if they are paid for it. I have already discussed Hume’s problem of induction and the occurrence of black swans. Here I introduce the scientific perpetrators. Recall that I have waged a war against the charlatanism of some prominent
…
a discussion of the LTCM events partook of a masquerade of science by adducing ad hoc explanations and putting the blame on a rare event (problem of induction: How did they know it was a rare event?). They spent their energy defending themselves rather than trying to make a buck with what they
A History of Western Philosophy
by Aaron Finkel · 21 Mar 1945 · 1,402pp · 369,528 words
that can be tested by observation. Usually the deduction is mathematical, and in this respect Bacon underestimated the importance of mathematics in scientific investigation. The problem of induction by simple enumeration remains unsolved to this day. Bacon was quite right in rejecting simple enumeration where the details of scientific investigation are concerned, for
Data Mining: Concepts, Models, Methods, and Algorithms
by Mehmed Kantardzić · 2 Jan 2003 · 721pp · 197,134 words
that are important guidelines in a practical implementation of data-mining techniques. Let us briefly explain two of these useful principles. First, when solving a problem of inductive learning based on finite information, one should keep in mind the following general commonsense principle: Do not attempt to solve a specified problem by indirectly
The Fabric of Reality
by David Deutsch · 31 Mar 2012 · 511pp · 139,108 words
to sequences of logical deductions from the evidence, what does it amount to? Why should we accept its conclusions? {58} This is known as the 'problem of induction'. The name derives from what was, for most of the history of science, the prevailing theory of how science works. The theory was that there
…
adopt our new set of theories in preference to the old set. That is why science, regarded as explanation-seeking and problem-solving, raises no 'problem of induction'. There is no mystery about why we should feel compelled tentatively to accept an explanation when it is the best explanation we can think of
…
solipsism The theory that only one mind exists and that what appears to be external reality is only a dream taking place in that mind. problem of induction Since scientific theories cannot be logically justified by observation, what does justify them? induction A fictitious process by which general theories were supposed to be
…
ibout the nature of reality. Yet these conclusions cannot be deduced by pure logic from the observations. So what makes them compelling? This is the 'problem of induction'. According to inductivism, ncientific theories are discovered by extrapolating the results of observations, and justified when corroborating observations are obtained. In fact, inductive reasoning is
…
impossible lo extrapolate observations unless one already has an explanatory Iramework for them. But the refutation of inductivism, and also the real solution of the problem of induction, depends on recognizing that science is a process not of deriving predictions from observations, but of finding explanations. We seek explanations when we encounter a
…
brief excursion into epistemology. << >> A Conversation About Justification (or 'David and the Crypto-inductivist') I think that I have solved a major philosophical problem: the problem of induction. Karl Popper As I explained in the Preface, this book is not primarily a defence of the fundamental theories of the four main strands; it
…
satisfied those criteria today possibly imply anything about what will happen if we rely on the theory tomorrow? This is the modern form of the 'problem of induction'. Most philosophers are now content with Popper's contention that new theories are not inferred from anything, but are merely hypotheses. They also accept that
…
they had succeeded, in the sense of constructing a scheme that could be followed successfully to create scientific knowledge, this would not have solved the problem of induction as it is nowadays understood. For in that case 'induction' would simply be another possible way of choosing theories, and the problem would remain of
…
why those theories should be a reliable basis for action. In other words, philosophers who worry about this 'problem of induction' are not inductivists in the old-fashioned sense. They do not try to obtain or justify any theories inductively. They do not expect the sky
…
an X-shaped gap in one's scheme of things and believing in X. Hence to fit in with the more sophisticated conception of the problem of induction, I wish to redefine the term 'inductivist' to mean someone who believes that the invalidity of inductive justification is a problem for the foundations of
…
as he sees it in an imaginary dialogue between Popper and several other philosophers, entitled 'Why Both Popper and Watkins Fail to Solve {143} the Problem of Induction'.* The setting is the top of the Eiffel Tower. One of the participants - the 'Floater' - decides to descend by jumping over the side instead of
…
what Popper has to say about induction, I have believed that he did indeed, as he claimed, solve the problem of induction. But few philosophers agree. Why? CRYPTO-INDUCTIVIST: Because Popper never addressed the problem of induction as we understand it. What he did was present a critique of inductivism. Inductivism said that there is an
…
inductivism had been {144} known almost since it was invented, and certainly since David Hume's critique of it in the early eighteenth century. The problem of induction is not how to justify or refute the principle of induction, but rather, taking for granted that it is invalid, how to justify any conclusion
…
the railing. CRYPTO-INDUCTIVIST: So in view of that, I repeat, the whole problem is to find what does justify the prediction. That is the problem of induction. DAVID: Well, that is the problem that Popper solved. CRYPTO-INDUCTIVIST: That's news to me, and I've studied Popper extensively. But anyway, what
…
be rejected too, for it is just another way of stating your theory. CRYPTO-INDUCTIVIST: Could it be that there is a solution of the problem of induction lurking here after all? Let me see. How does this insight about language change things? My argument relied upon an apparent symmetry between your position
…
difference between theories which make unexplained predictions and theories which don't, I must admit that this does look promising as a solution of the problem of induction. You seem to have discovered a way of justifying your future reliance on the theory of gravity, given only the past problem-situation (including past
…
between theories being justified by observations (as inductivists think) and being justified by argument. But Popper made no such distinction. And in regard to the problem of induction, he actually said that although future predictions of a theory cannot be justified, we should act as though they were! DAVID: I don't think
…
no principle of induction. There is no process of induction. No one ever uses them or anything like them. And there is no longer a problem of induction. Is that clear now? CRYPTO-INDUCTIVIST: Yes. Please excuse me for a few moments while I adjust my entire world-view. DAVID: To assist you
…
by some as yet unknown means. In either case there is a missing justification. I no longer suspect that this is the problem of induction in disguise. Nevertheless, having exploded the problem of induction, have we not revealed another fundamental problem, also concerning missing justification, beneath? DAVID: What justifies the principles of rationality? Argument, as usual
…
so blind? To think that I once nominated Popper for the Derrida Prize for Ridiculous {165} Pronouncements, while all the time he had solved the problem of induction! O mea culpa! God save us, for we have burned a saint! I feel so ashamed. I see no way out but to throw myself
…
the succession of paradigms is one illustration of this. More seriously, very few philosophers agree with Popper's claim that there is no longer a 'problem of induction' because we do not in fact obtain or justify theories from observations, but proceed by explanatory conjectures and refutations instead. It is not that many
…
defensive against obsolete theories. The debate between Popper and most of his critics was (as I said in Chapters 3 and 7) effectively about the problem of induction. Turing spent the last years of his life in effect defending the proposition that human brains do not operate by supernatural means. Everett left scientific
…
Weinberg, Steven ������pointlessness of the universe 346 ������unimportance of explanation 3�4 Wheeler, John Archibald 328 'Why Both Popper and Watkins Fail to Solve the Problem of Induction' (Worrall) 143�4 Wickramasinghe, Chandra 333 Wooters, William 278 world�view 62, 75, 83, 97, 159, 161, 169, 318�19, 321, 331, 335 ������Darwin and
Rationality: From AI to Zombies
by Eliezer Yudkowsky · 11 Mar 2015 · 1,737pp · 491,616 words
Being Wrong: Adventures in the Margin of Error
by Kathryn Schulz · 7 Jun 2010 · 486pp · 148,485 words
Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die
by Eric Siegel · 19 Feb 2013 · 502pp · 107,657 words
The Moral Landscape: How Science Can Determine Human Values
by Sam Harris · 5 Oct 2010 · 412pp · 115,266 words
Open Standards and the Digital Age: History, Ideology, and Networks (Cambridge Studies in the Emergence of Global Enterprise)
by Andrew L. Russell · 27 Apr 2014 · 675pp · 141,667 words
The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World
by Pedro Domingos · 21 Sep 2015 · 396pp · 117,149 words
Radical Uncertainty: Decision-Making for an Unknowable Future
by Mervyn King and John Kay · 5 Mar 2020 · 807pp · 154,435 words
The Language Instinct: How the Mind Creates Language
by Steven Pinker · 1 Jan 1994 · 661pp · 187,613 words
The Science of Language
by Noam Chomsky · 24 Feb 2012
The Invention of Science: A New History of the Scientific Revolution
by David Wootton · 7 Dec 2015 · 1,197pp · 304,245 words
The Information: A History, a Theory, a Flood
by James Gleick · 1 Mar 2011 · 855pp · 178,507 words
Surfaces and Essences
by Douglas Hofstadter and Emmanuel Sander · 10 Sep 2012 · 1,079pp · 321,718 words
The Stuff of Thought: Language as a Window Into Human Nature
by Steven Pinker · 10 Sep 2007 · 698pp · 198,203 words
Everything Is Predictable: How Bayesian Statistics Explain Our World
by Tom Chivers · 6 May 2024 · 283pp · 102,484 words
Safe Haven: Investing for Financial Storms
by Mark Spitznagel · 9 Aug 2021 · 231pp · 64,734 words
The Structure of Scientific Revolutions
by Thomas S. Kuhn and Ian Hacking · 1 Jan 1962 · 314pp · 91,652 words
Rationality: What It Is, Why It Seems Scarce, Why It Matters
by Steven Pinker · 14 Oct 2021 · 533pp · 125,495 words
The Filter Bubble: What the Internet Is Hiding From You
by Eli Pariser · 11 May 2011 · 274pp · 75,846 words
Think Twice: Harnessing the Power of Counterintuition
by Michael J. Mauboussin · 6 Nov 2012 · 256pp · 60,620 words
Understanding Sponsored Search: Core Elements of Keyword Advertising
by Jim Jansen · 25 Jul 2011 · 298pp · 43,745 words
Intertwingled: Information Changes Everything
by Peter Morville · 14 May 2014 · 165pp · 50,798 words
Fluke: Chance, Chaos, and Why Everything We Do Matters
by Brian Klaas · 23 Jan 2024 · 250pp · 96,870 words
How to Talk to a Science Denier: Conversations With Flat Earthers, Climate Deniers, and Others Who Defy Reason
by Lee McIntyre · 14 Sep 2021 · 407pp · 108,030 words
Science Fictions: How Fraud, Bias, Negligence, and Hype Undermine the Search for Truth
by Stuart Ritchie · 20 Jul 2020
The Rationalist's Guide to the Galaxy: Superintelligent AI and the Geeks Who Are Trying to Save Humanity's Future
by Tom Chivers · 12 Jun 2019 · 289pp · 92,714 words
Antifragile: Things That Gain From Disorder
by Nassim Nicholas Taleb · 27 Nov 2012 · 651pp · 180,162 words
Mind in Motion: How Action Shapes Thought
by Barbara Tversky · 20 May 2019 · 426pp · 117,027 words
Chaos Kings: How Wall Street Traders Make Billions in the New Age of Crisis
by Scott Patterson · 5 Jun 2023 · 289pp · 95,046 words
The Success Equation: Untangling Skill and Luck in Business, Sports, and Investing
by Michael J. Mauboussin · 14 Jul 2012 · 299pp · 92,782 words
The Fair Trade Scandal: Marketing Poverty to Benefit the Rich
by Ndongo Sylla · 21 Jan 2014 · 193pp · 63,618 words
The End of Growth: Adapting to Our New Economic Reality
by Richard Heinberg · 1 Jun 2011 · 372pp · 107,587 words
Beautiful Testing: Leading Professionals Reveal How They Improve Software (Theory in Practice)
by Adam Goucher and Tim Riley · 13 Oct 2009 · 351pp · 123,876 words
The Ascent of Money: A Financial History of the World
by Niall Ferguson · 13 Nov 2007 · 471pp · 124,585 words
The Personal MBA: A World-Class Business Education in a Single Volume
by Josh Kaufman · 2 Feb 2011 · 624pp · 127,987 words
Life Is Simple: How Occam's Razor Set Science Free and Shapes the Universe
by Johnjoe McFadden · 27 Sep 2021
Data Science for Business: What You Need to Know About Data Mining and Data-Analytic Thinking
by Foster Provost and Tom Fawcett · 30 Jun 2013 · 660pp · 141,595 words
Exploring the World of Lucid Dreaming
by Stephen Laberge, Phd and Howard Rheingold · 8 Feb 2015