Quantitative trading with futures and machine learning

Hedge funds employing quantitative methods represent a $1000b industry. Nowadays, machine learning methods are becoming increasingly popular among traders for dealing with more and better data. This talk will focus on algorithmic trading with futures contracts, addressing the challenge of applying machine learning methods to financial data. Developing a trading system can be view as tackling a supervised learning problem, and we will discuss one concrete example which compares the performance of different algorithms.

Friday, 23 June 2017, ore 14:30 — Aula Magna

Overview of bottomonium spectroscopy and prospects for the Belle2 experiment

Bottomonium physics has been a major success for the first generation of B-factories. Among the main achievements obtained, there are the discovery of several bottomonium states, such as etab(1S) and hb(1P,2P), and the observation of exotic resonances, namely Zb(10610) and Zb(10650). Moreover, important informations on bottomonium are given not only by new resonances, but also by the study of the transitions occurring between its states. In particular, the study of hadronic transitions among bottomonia, and their relative magnitude, can be used as a test bench for non-perturbative approaches to QCD. For instance, transitions through an eta meson, despite involving a heavy quark spin-symmetry violation, have been measured to have an unexpectedly enhanced branching fraction with respect to those through a dipion system. We summarise the heritage left to this field by the first generation of B-factories, enlightening also the open issues that are expected to be more extensively understood with the data collected by the Belle2 experiment.

Monday, 12 June 2017, ore 14:30 — Sala Wataghin

Computing jet-quenching and the transport coefficients of the Quark-Gluon plasma

Two of the main findings of the experimental program on ultrarelativistic heavy-ion collisions are that the produced medium appears to have a very small shear viscosity to entropy ratio and that jets traveling through this medium are strongly suppressed (jet-quenching). I will review these two aspects and the theoretical challenges in their description, focusing in particular on the effective kinetic theory approach, which allows to compute both the transport coefficients, such as the shear viscosity, and the medium modifications to jets. After introducing the main features of this framework, which requires the assumption of weak coupling, I will review the recent efforts in extending such approach to next-to-leading order, so that one can estimate the uncertainties that arise when extrapolating to realistic couplings.

Friday, 9 June 2017, ore 14:30 — Sala Wataghin

Neutrinoless double-beta decay searches with 76Ge

The search for neutrinoless double-beta decay might be the only window to observe lepton number violation and is therefore considered to be of highest relevance. The isotope 76Ge has historically been most important for this search and the ongoing experiment GERDA has the lowest background of all experiments in the field. The talk reviews the motivation, the current status of GERDA experiment and future programs.

Thursday, 1 June 2017, ore 14:30 — Sala Wataghin

The physics of Artificial Intelligence

In many applications Artificial Intelligence (AI) is reaching abilities that are comparable to those of human beings, if not better. These achievements were still impossible ten years ago and their technological follow-ups are rapidly impacting our societies. The main current tools of AI are artificial deep neural networks (nonlinear devices with hundreds of millions of connections) inspired by the human visual systems. The ability of deep artificial neural networks to learn efficiently from unorganized and huge data sets is inspiring researchers to think about artificial intelligence in an unprecedented way. Still the field is under many aspects heuristic and there is an urgent need for an in-depth theoretical comprehension. Learning is, in principle, a hard problem in deep neural networks. However, while standard techniques fail badly, in practice heuristic algorithms often find solutions with surprisingly good generalization properties. We discuss an explanation of this behavior in terms of a non-equilibrium statistical physics framework: we show that there exist rare regions of the optimization landscape that are both robust and accessible to out-of-equilibrium stochastic processes and that their existence is crucial to achieve good performance. Building on these analytical results, we provide a unified framework for understanding the success of the current learning algorithms and show how novel stochastic algorithmic schemes can be derived.

Thursday, 25 May 2017, ore 14:30 — Aula Magna

The search for neutrinoless double-beta decay

The discovery of neutrino mass suggests new physics, one that would violate lepton number. Lepton number violation in the early Universe is also a likely ingredient to explain why the cosmological baryon-antibaryon asymmetry, and ultimately life, have developed. In this talk I will explain why a hypothetical nuclear process, neutrinoless double beta decay, is considered to be the most promising way to search for lepton number violation, and to explore the origin of neutrino mass. Despite a 70-year long history and many null results, the experimental exploration of neutrinoless double beta decay is experiencing a golden age today. I will describe the experimental challenges being faced, and highlight some of the best approaches to overcome them. Among the many experimental techniques, and Nature permitting, who will win this healthy competition to discover neutrinoless double beta decay?

Friday, 19 May 2017, ore 14:30 — Sala Wataghin

Breaking Universality in Droplet and Nanoparticle Growth

Universality is a key concept in statistical physics. It has potent applications in the theory of critical phenomena and the description of growth processes. In the theory of critical phenomena the symmetries of the Hamiltonian uniquely select the set of critical exponents that govern power-law divergences of correlation functions and response properties. In the description of growth processes scaling theory has been adopted to predict a universal shape of particle- and droplet-size distributions. I will discuss the growth of droplets on a substrate (dew) and in the bulk (synthesis of nanoparticles). In the former case a critical exponent that characterizes the growth process is found to be non-universal. In the latter case the size distribution is not necessarily universal. We clarify why ripening processes and nanoparticle growth follow qualitatively different growth scenarios, and how this has been adopted to minimize particle-size dispersal. In an outlook I will briefly comment on non-universal aspects of the jamming critical point of granular shear flow, and formulate challenges for future work.

Friday, 12 May 2017, ore 14:30 — Aula A

From Quantum to Cosmos

The discovery in the late 1920s that our universe expands led the Belgian astronomer and priest Georges Lemaitre to conjecture it had a quantum origin. I sketch the modern framework on which a quantum approach to cosmology is based. In 1983 Hartle and Hawking implemented Lemaitre's vision and put forward the first concrete model to describe a quantum origin of the universe. Their model predicts our universe emerged with a period of inflation, a phase of rapid expansion which generates the seeds for a complex universe, starting from a natural beginning. However, a fuzzy quantum origin is bound to give rise to a multiverse of possible universes. I discuss some of the challenges associated with the development of a truly predictive multiverse cosmology that is falsifiable to observers within one of its universes. I close by sketching more recent developments aimed at firmly rooting quantum cosmology in fundamental high-energy physics.

Friday, 5 May 2017, ore 14:30 — Sala Wataghin

From heavy-ion to proton collisions: small systems get stranger

Heavy-ion collisions have traditionally been exploited to study the hot and dense deconfined phase of strongly-interacting matter: the Quark Gluon Plasma (QGP). In this context, proton-proton and proton-nucleus collisions have been used to set the reference and to study initial state effects. The high energy frontier at RHIC and LHC has revived a long-standing intriguing question: is it possible to produce QGP in small systems? The chemical composition of the system created in the collision is a key tool to single-out the conditions under which a QGP onset can occur. It has been extensively studied at the LHC in three collision systems and at different energies, leading to ground-breaking results that will be presented and discussed in view of possible phenomenological interpretations.

Friday, 28 April 2017, ore 14:30 — Aula F

The limits of Cosmology

One of our greatest challenges is understanding the nature of dark matter. Another is the origin of the structure of the universe, and in particular the formation of the galaxies. I will describe the progress of our search for dark matter by direct and indirect techniques. I will describe how the fossil radiation from the beginning of the universe, the cosmic microwave background, has provided a window for probing the initial conditions from which structure evolved and seeded the formation of the galaxies, and the outstanding issues that remain to be resolved. I will address our optimal choice of future strategy in order to make further progress on understanding our cosmic origins.

Friday, 21 April 2017, ore 14:30 — Sala Wataghin