|
Calendar of Physics Talks Vienna
| Boundary actions for three dimensional gravity |
| Speaker: | Wout Merbis (ULB) |
| Abstract: | Gravity in three space-time dimensions (3D) admits a Chern-Simons formulation, which has several merits. It highlights the topological properties of gravity in 3D, can facilitate quantization, and is inherently related to holography due to the appearance of edge modes on manifolds with a (conformal) boundary. In this talk we will discuss how three dimensional gravity on manifolds with a boundary may be reduced to a theory on the boundary. We show the boundary dynamics is determined by the asymptotic symmetry group and the boundary theory is related to the geometric action on the coadjoint orbits of this asymptotic symmetry group. We will then discuss several application of this reduction in AdS_3 and in flat space. |
| Date: | Thu, 13.02.2020 |
| Time: | 16:00 |
| Duration: | 60 min |
| Location: | TU Wien, Wiedner Hauptstraße 10, Yellow Area, Seminar Room |
| Contact: | Céline Zwikel |
| Is model transparency good or bad for scientific applications of deep learning? |
| Speaker: | Cameron Buckner (University of Houston) |
| Abstract: | As one might expect from talk by a philosopher, the answer to the title question is a big “it depends”; it depends on what we mean by ‘transparency’, it depends on the uses to which the model’s verdicts will be put (especially their epistemological, ethical, and legal contexts), and the things on which it depends are much more complicated than we very recently supposed. In this talk, I begin by briefly reviewing the leading theories about why deep learning neural networks are often more efficient and accurate than alternative methods in scientific data analysis. I will then raise a cluster of black box/interpretability challenges to these methods, and taxonomize the different ways that interpretability might be understood here. In particular, I will use the problem of adversarial examples—artificially created data points that seem to dramatically fool deep learning neural networks,... |
| Date: | Fri, 14.02.2020 |
| Time: | 14:30 |
| Duration: | 60 min |
| Location: | Christian Doppler lecture hall, Faculty of Physics, University of Vienna |
| Contact: | VDS Physics |
| Machine Learning in Physics: Panel Discussion |
| Speaker: | Torsten Möller, Peter Wirnsberger, Stefan Thurner, Cameron Buckner |
| Date: | Fri, 14.02.2020 |
| Time: | 15:30 |
| Duration: | 60 min |
| Location: | Christian Doppler lecture hall, Faculty of Physics, University of Vienna |
| Contact: | VDS Physics |
|