Fondements de la mécanique quantique et information quantique

Les fondements de la mécanique quantique sont depuis toujours au centre de débats philosophiques. Depuis les années 1990 l’information quantique a renouvelé l’intérêt pour ces débats parmi les physiciens et y a ajouté une nouvelle perspective. Ce colloque réunit des experts de ces débats travaillant en France et à l’étranger. Elle s’inscrit dans le cadre d’une forte activité récente dans le domaine des fondements de la physique en région parisienne, notamment dans un projet financé par l’ANR réunissant l’IHPST, le CREA et le CEA

Programme

10h : Armond Duwell, Department of Philosophy, University of Montana
Cluster state quantum computers and accounts of computation
Abstract: Recently Piccinini (2007) has advocated a set of desiderata for any account of computation to satisfy:

1. Objectivity: An account of computation should make it a matter of fact which systems compute a function.
2. Explanation: An account should explain computer's abilities in virtue of the computations they perform.
3. The right things compute, and wrong things do not.
4. Miscomputation: An account of computation should reveal how miscomputation is possible.
5. Taxonomy: An account of computation should have the resources to explain different capacities of different computers.

Piccinini has argued that his functional account of computation that relies heavily on mechanistic explanation of computational processes satisfies this set of criteria.

I will challenge Piccinini via an examination of one-way or cluster state quantum computers. I will argue that Piccinini's account is too closely linked to standard models of computation. It recovers our everyday intuitions, but provides us little reassurance that the account exhausts possibilities for computation.

My paper will primarily focus on Piccinini's explication of desideratum 2. which focuses on the proper form of a computational explanation. I will argue that cluster state quantum computers cannot be explained as Piccinini suggests they ought to be if they are genuine computers. I.e. they must manipulate strings of digits which initially enter into the computational device, persist, and then exit. I will suggest that a less restrictive account along the lines of Copeland's (1996) is called for and argue that it is not in conflict with a proper construal of the above desiderata.

11h : Soazig Le Bihan, Department of Philosophy, University of Montana
Bell-type correlations: the hidden causal picture re-assessed
Abstract: Bell-type phenomena are characterized by striking correlations between spatially separated outcome events. We typically take correlations at the phenomenal level to be indicative of an underlying causal structure. Bell-type correlations are no exception. The received view is that failure of parameter independence is indicative of an underlying causal relationship but failure of outcome independence is not. The problem with the received view is that it does not explicitly articulate a notion of causation that can support that view. Now, there are currently at least three viable theories of causation. I will argue that the causal structure of Bell-type phenomena varies according to which theory of causation one considers. My conclusions will that: (1) none of the theories of causation considered support the received view, if it is understood as a strong metaphysical thesis; but (2) the received view can be rigorously supported if its claims are restricted to the empirical level.

14h30 : Guido Bacciagaluppi, Department of Philosophy, University of Aberdeen
Decohering away the tails problem
Abstract: The spontaneous collapse approach to quantum mechanics is often criticised on the basis that it supposedly does not yield full localisation of macroscopic objects. The most serious version of this so-called 'tails' problem is that proposed by Everettians, who argue that the spontaneous collapse does not get rid of other 'worlds'. I suggest that, instead, the combination of the process of decoherence (that is normally invoked by Everettians for defining worlds) with the collapse mechanism will not lead to the formation of other worlds, so that the tails problem is dissolved.

15h30 : Alexei Grinbaum, LARSIM, CEA TBA

16h30 : Commentaires par Alexandre Guay, Université de Bourgogne et Thomas Boyer, IHPST

Répondants : Alexandre Guay, Université de Bourgogne et Thomas Boyer, IHPST