Dear all,

Those at Nikhef, please attend this seminar. Gilles was working at CERN previously and did a lot of developments together with LHCb people.

Cheers,

Patrick



-------- Forwarded Message --------
Subject: [Colloquium] Nikhef colloquium announcement
Date: Wed, 27 Sep 2017 08:36:45 +0200
From: Wijnanda de Haan <wdehaan@nikhef.nl>
To: Colloquium <colloquium@nikhef.nl>



                                                Nikhef colloquium
                                                ------------------

title    : "Teaching machines to discover particles"
speaker  : Dr. Gilles Louppe (NYU)
date     : Friday 29th September 2017
place    : H331 (Nikhef)
time     : 11:00

summary:
Across many fields of science, computer simulators are used to describe complex
data generation processes. These simulators relate observations to the
parameters of an underlying theory or mechanistic model. In most cases, they are
specified as procedural implementations of forward stochastic processes
involving latent variables. Rarely do these simulators admit a tractable density
or likelihood function, thereby making inference difficult. The prevalence and
significance of this problem has motivated an active research effort in
so-called likelihood-free inference algorithms.

In this talk, we will discuss how likelihood-free inference has been carried for
decades in particle physics and then expand to new methods from machine
learning, including likelihood-ratio estimation through supervised learning [1]
and adversarial variational optimization [2].

Building upon these algorithms, we will then discuss how artificial intelligence
can enable the automation of the scientific method through the synergy of
three powerful techniques:
- Generic likelihood-free inference engines that enable statistical
  inference​ ​on​ the​ ​ parameters​​ of​ a​ theory​​ that​​ are​  implicitly​ ​
  defined​ ​by​ a​ simulator.
- Sequential design algorithms (e.g., Bayesian optimization) that balance
  exploration and exploitation​ ​ to​ ​ efficiently​ ​ optimize​ ​ an​ ​ expensive​ ​
  black​ ​ box​ ​ function.
- Workflows that encapsulate scientific pipelines and extend the scope
  from reproducibility to​​ reusability.​

[1] https://arxiv.org/abs/1506.02169
[2] https://arxiv.org/abs/1707.07113
The Nikhef colloquiumcommittee Robert Fleischer en Panos Christakoglou For the full list of (planned) colloquia please see the agenda http://indico.nikhef.nl/categoryDisplay.py?categId=5 -- Secretariaat Nikhef Science Park 105 (v/h Kruislaan 409), 1098 XG Amsterdam Postbus 41882, 1009 DB Amsterdam T 0031 - 20 - 592 5169 F 0031 - 20 - 592 5054 E jberger@nikhef.nl (aanwezig di t/m do) _______________________________________________ Colloquium mailing list Colloquium@nikhef.nl https://mailman.nikhef.nl/mailman/listinfo/colloquium