Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:0711.2559 · Physical review. D. Particles, fields, gravitation, and cosmology/Physical review. D. Particles and fields

Prediction and explanation in the multiverse

Jaume Garriga, Alexander Vilenkin

Published Feb 26, 2008• 70 citations
arXivHTMLPDF

Abstract

Probabilities in the multiverse can be calculated by assuming that we are typical representatives in a given reference class. But is this class well defined? What should be included in the ensemble in which we are supposed to be typical? There is a widespread belief that this question is inherently vague, and that there are various possible choices for the types of reference objects which should be counted in. Here we argue that the ``ideal'' reference class (for the purpose of making predictions) can be defined unambiguously in a rather precise way, as the set of all observers with identical information content. When the observers in a given class perform an experiment, the class branches into subclasses who learn different information from the outcome of that experiment. The probabilities for the different outcomes are defined as the relative numbers of observers in each subclass. For practical purposes, wider reference classes can be used, where we trace over all information which is uncorrelated to the outcome of the experiment, or whose correlation with it is beyond our current understanding. We argue that, once we have gathered all practically available evidence, the optimal strategy for making predictions is to consider ourselves typical in any reference class we belong to, unless we have evidence to the contrary. In the latter case, the class must be correspondingly narrowed.

DOI
10.1103/physrevd.77.043526

Related papers

Quantum Theory and Measurement

Wojciech H. Zurek

We sketch a quantum-mechanical framework for the universe as a whole. \nWithin that framework we propose a program for describing the ultimate \norigin in quantum cosmology of the "quasiclassical domain" of familiar experience and for characterizing the process of measurement. Predictions \nin quantum mechanics are made from probabilities for sets of alternative \nhistories. Probabilities (approximately obeying the rules of probability theory) can be assigned only to sets of histories that approximately decohere. \nDecoherence is defined and the mechanism of decoherence is reviewed. Decoherence requires a sufficiently coarse-grained description of alternative \nhistories of the universe. A quasiclassical domain consists of a branching set \nof alternative decohering histories, described by a coarse graining that is, in \nan appropriate sense, maximally refined consistent with decoherence, with \nindividual branches that exhibit a high level of classical correlation in time. \nWe pose the problem of making these notions precise and quantitative. A \nquasiclassical domain is emergent in the universe as a consequence of the \ninitial condition and the action function of the elementary particles. It is \nan important question whether all the quasiclassical domains are roughly \nequivalent or whether there are various essentially inequivalent ones. A \nmeasurement is a correlation with variables in a quasiclassical domain. An "observer" (or information gathering and utilizing system) is a complex \nadaptive system that has evolved to exploit the relative predictability of a \nquasiclassical domain, or rather a set of such domains among which it cannot discriminate because of its own very coarse graining. We suggest that \nresolution of many of the problems of interpretation presented by quantum \nmechanics is to be accomplished, not by further scrutiny of the subject as it \napplies to reproducible laboratory situations, but rather by an examination \nof alternative histories of the universe, stemming from its initial condition, \nand a study of the problem of quasiclassical domains.

CRC Press eBooks•Mar 8, 2018•61 citations

Multiverse interpretation of quantum mechanics

Raphael Bousso A, Leonard Susskind

We argue that the many worlds of quantum mechanics and the many worlds of the multiverse are the same thing, and that the multiverse is necessary to give exact operational meaning to probabilistic predictions from quantum mechanics. Decoherence---the modern version of wave-function collapse---is subjective in that it depends on the choice of a set of unmonitored degrees of freedom, the environment. In fact decoherence is absent in the complete description of any region larger than the future light cone of a measurement event. However, if one restricts to the causal diamond---the largest region that can be causally probed---then the boundary of the diamond acts as a one-way membrane and thus provides a preferred choice of environment. We argue that the global multiverse is a representation of the many worlds (all possible decoherent causal diamond histories) in a single geometry. We propose that it must be possible in principle to verify quantum-mechanical predictions exactly. This requires not only the existence of exact observables but two additional postulates: a single observer within the Universe can access infinitely many identical experiments; and the outcome of each experiment must be completely definite. In causal diamonds with a finite surface area, holographic entropy bounds imply that no exact observables exist, and both postulates fail: experiments cannot be repeated infinitely many times; and decoherence is not completely irreversible, so outcomes are not definite. We argue that our postulates can be satisfied in hats (supersymmetric multiverse regions with vanishing cosmological constant). We propose a complementarity principle that relates the approximate observables associated with finite causal diamonds to exact observables in the hat.

arXiv:1105.3796•Physical review. D. Particles, fields, gravitation, and cosmology/Physical review. D. Particles and fields•Feb 6, 2012•155 citations

What is physics? The individual and the universal, and seeing past the noise

A. R. P. Rau

Along with weaving together observations, experiments, and theoretical constructs into a coherent mesh of understanding of the world around us, physics over its past five centuries has continuously refined the base concepts on which the whole framework is built. In quantum physics, first in non-relativistic mechanics and later in quantum field theories, even familiar concepts of position, momentum, wave or particle, are derived constructs from the classical limit in which we live but not intrinsic to the underlying physics. Most crucially, the very idea of the individual, whether an object or an event, distinguished only in a mere label of identity from others identical to it in all the physics, exists only as an approximation, not an element of underlying reality. Failure to recognize this and seeking alternative explanations in many worlds or multiverses leads only to incoherent logic and incorrect physics. As an example, in a physical system such as an atom in a particular state, physics deals with the universal system of all such atoms but makes no meaningful prediction of the position of an electron or the time of decay of any specific atom. Those are incidental, entirely random among all possible positions and times, even while physics makes very precise predictions for the distribution of the outcomes in measurements on atoms in that state. Physics deals with the universal, not the individual.

arXiv:1605.03106•Physics Essays•Mar 16, 2017•3 citations

Evidence and Uncertainty in Everett’s Multiverse

Paul Tappenden

How does it come about then, that great scientists such as Einstein, Schrödinger and De Broglie are nevertheless dissatisfied with the situation? Of course, all these objections are levelled not against the correctness of the formulae, but against their interpretation. […] The lesson to be learned from what I have told of the origin of quantum mechanics is that probable refinements of mathematical methods will not suffice to produce a satisfactory theory, but that somewhere in our doctrine is hidden a concept, unjustified by experience, which we must eliminate to open up the road. (Born [1954], pp. 8, 11) It is truly surprising how little difference all this makes. Most physicists use quantum mechanics every day in their working lives without needing to worry about the fundamental problem of its interpretation. (Weinberg [1992], p. 66) I endorse the view that it may be of no relevance to the acceptability of the Everett interpretation of quantum mechanics as a physical theory whether or not an informed observer can be uncertain about the outcome of a quantum measurement prior to its having occurred. However, I suggest that the very possibility of post-measurement, pre-observation uncertainty has an essential role to play in both confirmation theory and decision theory in a branching universe. This is supported by arguments which do not appeal to van Fraassen’s Reflection Principle. 1. Dendritic Unitary Evolution 2. The Born–Vaidman Rule 3. Greaves on Uncertainty 4. Retroactive Counterfactual Uncertainty 5. Odds and Ends

The British Journal for the Philosophy of Science•Jun 11, 2010•63 citations

A discrete, finite multiverse

McKenzie, Alan

The Many Worlds Interpretation (MWI) famously avoids the issue of wave function collapse. Different MWI trees representing the same quantum events can have different topologies, depending upon the observer. However, they are all isomorphic to the group of block universes containing all of the outcomes of all of the events, and so, in that sense, the group of block universes is a more fundamental representation. Different branches of the MWI tree, representing different universes in MWI, ultimately share the same quantum state in a common ancestor branch. This branching topology is incompatible with that of the Minkowski block universe; the resolution is to replace the branches with discrete, parallel block universes, each of which extends from the trunk to the outermost twigs. The number of universes in a branch is proportional to its thickness which, in turn, depends upon the absolute square of the probability amplitude for the state in that branch. Every quantum event may be represented by a kernel of universes, which is the smallest group of universes that will reproduce the quantum probabilities of the outcomes of that event. By considering the ratios of the probabilities of the outcomes of any event, it can be shown that the number of universes in every kernel must finite, as must be the total number of universes in the multiverse. Further, every universe in the multiverse must be finite both in space and time. Another consequence is that quantum probabilities must be rational, which suggests that quantum mechanics is only an approximation to a discrete theory. A corollary is that not every conceivable universe exists in the multiverse, no doubt to the disappointment of those who enjoy alternate-history novels.

arXiv:1609.04050•Sep 10, 2016•2 citations

Typical Quantum States of the Universe are Observationally Indistinguishable

Keming Chen, Roderich Tumulka

We establish three impossibility results regarding our knowledge of the quantum state of the universe. Suppose the universal quantum state is a typical unit vector in a high-dimensional subspace $\mathscr{H}_0$ of Hilbert space $\mathscr{H}$, such as the low-entropy subspace defined by the Past Hypothesis. We show that: (1) Any particular observation is incapable of identifying the universal state vector in $\mathscr{H}_0$ or substantially reducing the set of possibilities. In other words, the overwhelming majority of possible state vectors are observationally indistinguishable from each other. (2) For any reasonably probable measurement outcome and for most pairs of vectors in $\mathscr{H}_0$, that outcome will not appreciably favor one vector over the other. (3) Bayesian updating on any measurement result, unless it is extraordinarily improbable, has a negligible effect on the initial uniform probability distribution over the states in $\mathscr{H}_0$. These findings represent the most stringent epistemic constraints known for a quantum universe and are derived from a typicality theorem in quantum statistical mechanics. We close by considering how theoretical considerations beyond empirical evidence might inform our understanding of this fact and our knowledge of the universal quantum state.

arXiv:2410.16860•The British Journal for the Philosophy of Science•Jan 22, 2026•1 citation
Probabilities in the multiverse can be calculated by assuming that we are typical representatives in a given reference class. But is this class well defined? What should be included in the ensemble in which we are supposed to be typical? There is a widespread belief that this question is inherently vague, and that there are various possible choices for the types of reference objects which should be counted in. Here we argue that the ``ideal'' reference class (for the purpose of making predictions) can be defined unambiguously in a rather precise way, as the set of all observers with identical information content. When the observers in a given class perform an experiment, the class branches into subclasses who learn different information from the outcome of that experiment. The probabilities for the different outcomes are defined as the relative numbers of observers in each subclass. For practical purposes, wider reference classes can be used, where we trace over all information which is uncorrelated to the outcome of the experiment, or whose correlation with it is beyond our current understanding. We argue that, once we have gathered all practically available evidence, the optimal strategy for making predictions is to consider ourselves typical in any reference class we belong to, unless we have evidence to the contrary. In the latter case, the class must be correspondingly narrowed.
We sketch a quantum-mechanical framework for the universe as a whole. \nWithin that framework we propose a program for describing the ultimate \norigin in quantum cosmology of the "quasiclassical domain" of familiar experience and for characterizing the process of measurement. Predictions \nin quantum mechanics are made from probabilities for sets of alternative \nhistories. Probabilities (approximately obeying the rules of probability theory) can be assigned only to sets of histories that approximately decohere. \nDecoherence is defined and the mechanism of decoherence is reviewed. Decoherence requires a sufficiently coarse-grained description of alternative \nhistories of the universe. A quasiclassical domain consists of a branching set \nof alternative decohering histories, described by a coarse graining that is, in \nan appropriate sense, maximally refined consistent with decoherence, with \nindividual branches that exhibit a high level of classical correlation in time. \nWe pose the problem of making these notions precise and quantitative. A \nquasiclassical domain is emergent in the universe as a consequence of the \ninitial condition and the action function of the elementary particles. It is \nan important question whether all the quasiclassical domains are roughly \nequivalent or whether there are various essentially inequivalent ones. A \nmeasurement is a correlation with variables in a quasiclassical domain. An "observer" (or information gathering and utilizing system) is a complex \nadaptive system that has evolved to exploit the relative predictability of a \nquasiclassical domain, or rather a set of such domains among which it cannot discriminate because of its own very coarse graining. We suggest that \nresolution of many of the problems of interpretation presented by quantum \nmechanics is to be accomplished, not by further scrutiny of the subject as it \napplies to reproducible laboratory situations, but rather by an examination \nof alternative histories of the universe, stemming from its initial condition, \nand a study of the problem of quasiclassical domains.
We argue that the many worlds of quantum mechanics and the many worlds of the multiverse are the same thing, and that the multiverse is necessary to give exact operational meaning to probabilistic predictions from quantum mechanics. Decoherence---the modern version of wave-function collapse---is subjective in that it depends on the choice of a set of unmonitored degrees of freedom, the environment. In fact decoherence is absent in the complete description of any region larger than the future light cone of a measurement event. However, if one restricts to the causal diamond---the largest region that can be causally probed---then the boundary of the diamond acts as a one-way membrane and thus provides a preferred choice of environment. We argue that the global multiverse is a representation of the many worlds (all possible decoherent causal diamond histories) in a single geometry. We propose that it must be possible in principle to verify quantum-mechanical predictions exactly. This requires not only the existence of exact observables but two additional postulates: a single observer within the Universe can access infinitely many identical experiments; and the outcome of each experiment must be completely definite. In causal diamonds with a finite surface area, holographic entropy bounds imply that no exact observables exist, and both postulates fail: experiments cannot be repeated infinitely many times; and decoherence is not completely irreversible, so outcomes are not definite. We argue that our postulates can be satisfied in hats (supersymmetric multiverse regions with vanishing cosmological constant). We propose a complementarity principle that relates the approximate observables associated with finite causal diamonds to exact observables in the hat.
Along with weaving together observations, experiments, and theoretical constructs into a coherent mesh of understanding of the world around us, physics over its past five centuries has continuously refined the base concepts on which the whole framework is built. In quantum physics, first in non-relativistic mechanics and later in quantum field theories, even familiar concepts of position, momentum, wave or particle, are derived constructs from the classical limit in which we live but not intrinsic to the underlying physics. Most crucially, the very idea of the individual, whether an object or an event, distinguished only in a mere label of identity from others identical to it in all the physics, exists only as an approximation, not an element of underlying reality. Failure to recognize this and seeking alternative explanations in many worlds or multiverses leads only to incoherent logic and incorrect physics. As an example, in a physical system such as an atom in a particular state, physics deals with the universal system of all such atoms but makes no meaningful prediction of the position of an electron or the time of decay of any specific atom. Those are incidental, entirely random among all possible positions and times, even while physics makes very precise predictions for the distribution of the outcomes in measurements on atoms in that state. Physics deals with the universal, not the individual.
How does it come about then, that great scientists such as Einstein, Schrödinger and De Broglie are nevertheless dissatisfied with the situation? Of course, all these objections are levelled not against the correctness of the formulae, but against their interpretation. […] The lesson to be learned from what I have told of the origin of quantum mechanics is that probable refinements of mathematical methods will not suffice to produce a satisfactory theory, but that somewhere in our doctrine is hidden a concept, unjustified by experience, which we must eliminate to open up the road. (Born [1954], pp. 8, 11) It is truly surprising how little difference all this makes. Most physicists use quantum mechanics every day in their working lives without needing to worry about the fundamental problem of its interpretation. (Weinberg [1992], p. 66) I endorse the view that it may be of no relevance to the acceptability of the Everett interpretation of quantum mechanics as a physical theory whether or not an informed observer can be uncertain about the outcome of a quantum measurement prior to its having occurred. However, I suggest that the very possibility of post-measurement, pre-observation uncertainty has an essential role to play in both confirmation theory and decision theory in a branching universe. This is supported by arguments which do not appeal to van Fraassen’s Reflection Principle. 1. Dendritic Unitary Evolution 2. The Born–Vaidman Rule 3. Greaves on Uncertainty 4. Retroactive Counterfactual Uncertainty 5. Odds and Ends
The Many Worlds Interpretation (MWI) famously avoids the issue of wave function collapse. Different MWI trees representing the same quantum events can have different topologies, depending upon the observer. However, they are all isomorphic to the group of block universes containing all of the outcomes of all of the events, and so, in that sense, the group of block universes is a more fundamental representation. Different branches of the MWI tree, representing different universes in MWI, ultimately share the same quantum state in a common ancestor branch. This branching topology is incompatible with that of the Minkowski block universe; the resolution is to replace the branches with discrete, parallel block universes, each of which extends from the trunk to the outermost twigs. The number of universes in a branch is proportional to its thickness which, in turn, depends upon the absolute square of the probability amplitude for the state in that branch. Every quantum event may be represented by a kernel of universes, which is the smallest group of universes that will reproduce the quantum probabilities of the outcomes of that event. By considering the ratios of the probabilities of the outcomes of any event, it can be shown that the number of universes in every kernel must finite, as must be the total number of universes in the multiverse. Further, every universe in the multiverse must be finite both in space and time. Another consequence is that quantum probabilities must be rational, which suggests that quantum mechanics is only an approximation to a discrete theory. A corollary is that not every conceivable universe exists in the multiverse, no doubt to the disappointment of those who enjoy alternate-history novels.
We establish three impossibility results regarding our knowledge of the quantum state of the universe. Suppose the universal quantum state is a typical unit vector in a high-dimensional subspace H0\mathscr{H}_0H0​ of Hilbert space H\mathscr{H}H, such as the low-entropy subspace defined by the Past Hypothesis. We show that: (1) Any particular observation is incapable of identifying the universal state vector in H0\mathscr{H}_0H0​ or substantially reducing the set of possibilities. In other words, the overwhelming majority of possible state vectors are observationally indistinguishable from each other. (2) For any reasonably probable measurement outcome and for most pairs of vectors in H0\mathscr{H}_0H0​, that outcome will not appreciably favor one vector over the other. (3) Bayesian updating on any measurement result, unless it is extraordinarily improbable, has a negligible effect on the initial uniform probability distribution over the states in H0\mathscr{H}_0H0​. These findings represent the most stringent epistemic constraints known for a quantum universe and are derived from a typicality theorem in quantum statistical mechanics. We close by considering how theoretical considerations beyond empirical evidence might inform our understanding of this fact and our knowledge of the universal quantum state.