Quantum Foundations of a Classical Universe Conference - Participant interests

Participant Interests

Charles Bennett

I am interested in learning about the paradoxes of measure and observership, whether they can be mitigated by a less experiential and more algorithmic-informational definition of the observer, and whether a finite amount of fine tuning suffices to generate unboundedly growing complexity in the sense of logical depth. In the theory of fault-tolerant computation, stochastic models with finite fine tuning (i.e. having positive measure in the parameter space of their initial conditions and local transition probabilities) have been shown (cf. Gacs arXiv:math/0003117) to be "self-organizing" in low-dimensional Euclidean space, producing structures that are complex almost everywhere, instead of merely infinitely often as Boltzmann imagined. Can analogous phenomena occur in the non-Euclidean manifolds relevant to cosmology?

Sean Carroll

What I'd like to like to hear most from others is quite vague and general:

  • Given a wave function, how many "branches" does it have?
  • How are they defined, if they are at all?
  • To what extent is other information (system/environment split, sets of macroscopic observables) necessary for addressing such questions? '

James Hartle

In no particular order I would like to hear:

  • Jess Riedel on quantum darwinism and decoherent histories.
  • Scott Arronson on measures of complexity.
  • Charlie Bennett on what he thinks the important questions are. '
  • Wojtek Zurek on defining classicality without histories.
  • Sean Carroll and Don Page in a debate on Born's rule. Sean has written on how to derive it. Don has written about how it must be replaced.
  • Adrian Kent on path integrals and reality and the connection of his work to other similar ventures.
  • Stefan Leichenauer and Ken Olum on the measure problem in cosmology and whether this requires a modification of quantum mechanics.

Stefan Leichenauer

Questions for the group:

  • Is it possible to have multiple classical realities emerge from the same quantum system, depending on how you look at it (i.e., what subsystem you choose to call the observer)?
  • In standard treatments of decoherence, definite Hilbert space factors are assigned to an observer, environment, and so on at the beginning. But if we start from with a fundamental description of the whole system and are asked to identify the observer, etc., based on emergent classicality, how do we go about doing this? For instance, in AdS/CFT we may have a situation where the gravitational description of the system looks classical but the underlying quantum degrees of freedom are most easily presented as a strongly-coupled quantum field theory, and the dictionary relating the two descriptions is complicated. Is there a way to "find" classicality that may be hidden in such a system?

Ken Olum

The things I would most like to hear about, in 5 words each, are: "Can we live without anthropics?" and "Can we live with anthropics?".

Don Page

Here are some questions and subjects I would be interested in:

  • How much mindreading occurs?
  • What is the relationship between consciousness and decoherence?
  • If Sensible Quantum Mechanics or Mindless Sensationalism is correct, in the approximation of local quantum field theory, are the operators for a human sentient experience given by a weighted integral ver spacetime of operators localized in space to regions of size comparable to that of a human brain? To a significantly smaller size (say of the size of some smaller structure within the human brain)? How much localization in time does the operator have before being integrated over spacetime? Can it be essentially instantaneous (perhaps with an extent of roughly the Planck time), or does it need a much larger spread over time (e.g., of a magnitude of the order of 0.1 seconds, or even conceivably of the order of a human lifetime or perhaps even of the order of the species lifetime of humans or of the time for life to last on a planet, though it is very hard for me to see why any of these very long times should be needed)?
  • Are there plausible schemes for a nonlinear quantum-consciousness connection?
  • Is the quantum-consciousness connection the only thing that distinguishes a quantum state of the universe from unitarily related states, or are there other preferred operators (e.g., certain localized operators if local quantum field theory were valid, but it probably is not in quantum gravity) such that if they are kept fixed, the quantum state is not invariant under unitary transformations?
  • What, if any, is an approximate relationship between the measure of a sentient experience (which if normalized could be taken to be the probability of that experience and used as the likelihood of a theory if the experience were considered to be the observation used to test the theory) and some measure of the size or complexity of the observer?
  • When can a quantum state, pure or mixed, be said to be classical? Can this definition be made statically, or are dynamical consideration needed?
  • When can fluctuations such as Boltzmann brains be said to occur in a quantum system, such a thermal state, whose density matrix is time-independent?

[Further details].

Jess Riedel

With regards to self-locating uncertainty:

  • What is the core conflict in the question of whether one must use a uniform "xerographic" distribution over all compatible observers, or whether one may modify it with a Bayesian update?
  • What are the key thought experiments on both sides?
  • How does this compare to opinions from the academic philosophy community (e.g. Bostrom's "Anthropic Bias")?

More narrowly, and towards my own work, I am interested to what extent the requirement of redundant records might produce the wrong branch structure:

  • Are there any unambiguously classical variables that aren't recorded redundantly in many spatially separated regions?
  • Conversely, could there be such redundantly recorded variables that should not be considered classical?
  • Given thermodynamics, is there a redundantly recorded variable that is at risk of "recohering"?
  • Is there any reason to think that the uniqueness of redundantly recorded records (arXiv:1310.4473) will not be robust when extended to slightly imperfect records, or to small changes in the tensor-product structure defining subsystems?

Mark Srednicki

What I would most like to hear about: What is the criterion for "instantiation" of an event in a quantum system?

Consider a system in a state |psi>. We then apply a time-ordered sequence of projectors P_{a_i}(t_i) corresponding to some observable(s) being in some range at each time. We then compute the norm of the projected state. Many physicists (generally anyone not steeped in quantum foundations) will say that this norm gives the probability that this sequence of observables "happened"; that the system has this probability to have "experienced" this history. On the other hand, many coming to this conference would say that this is only true if this history is part of a decohering set.

Which is right (if either), and why? Is this even a scientific question that can, in principle, be decided? Or is it stuck in the murk of interpretation? <xsscleaned_tag>