BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Multi-agent paradoxes in quantum theory &amp\; Discovering physica
 l principles with neural networks(2 mini talks) - Nuriya Nurgalieva &amp\;
  Lídia del Río
DTSTART:20181102T120000Z
DTEND:20181102T130000Z
UID:TALK113158@talks.cam.ac.uk
CONTACT:Johannes Bausch
DESCRIPTION:*Inadequacy of multi-agent logic in quantum settings*\n\n_Nuri
 ya Nurgalieva (ETH Zurich)\, 25min_\n\nWe probe basic logical principles i
 n multi-agent quantum settings\, where parties are equipped with quantum m
 emories\, like in a network of quantum computers. A good testing ground is
  provided by Frauchiger-Renner thought experiment\, where agents argue abo
 ut the outcomes of each other's measurements. Using this example\, we show
  that classical logic cannot be applied to general settings where quantum 
 experiments are conducted\, a result that impacts quantum cryptography and
  communication theory. We propose ways to weaken logical axioms as to make
  them robust under quantum settings\, and conjecture that doing so would r
 equire parties to keep exponentially large memories. Finally\, we find a m
 issing assumption in the Frauchiger-Renner result.\n\nBased on https://arx
 iv.org/abs/1804.01106 joint with Lídia del Rio\n\n\n\n*Discovering physic
 al concepts with neural networks*\n\n_Lídia del Rio (ETH Zurich)\, 25min_
 \n\nThe formalism of quantum physics is built upon that of classical mecha
 nics. In principle\, considering only experimental data without prior know
 ledge could lead to an alternative quantum formalism without conceptual is
 sues like the measurement problem. As a first step towards finding such an
  alternative\, we introduce a neural network architecture that models the 
 physical reasoning process and can be used to extract physical concepts fr
 om experimental data in an unbiased way. We apply the neural network to a 
 variety of simple physical examples in classical and quantum mechanics\, l
 ike damped pendulums\, two-particle collisions\, and qubits. The network f
 inds the physically relevant parameters\, exploits conservation laws to ma
 ke predictions\, and can be used to gain conceptual insights. For example\
 , given a time series of the positions of the Sun and Mars as observed fro
 m Earth\, the network discovers the heliocentric model of the solar system
  - that is\, it encodes the data into the angles of the two planets as see
 n from the Sun.\n\nBased on http://arxiv.org/abs/1807.10300 joint with Rab
 an Iten\, Tony Metger\, Henrik Wilming and Renato Renner\n
LOCATION:MR21\, Centre for Mathematical Sciences\, Wilberforce Road\, Camb
 ridge
END:VEVENT
END:VCALENDAR