Barry Lam

Associate Professor of Philosophy at Vassar College and host of the philosophy podcast Hi-Phi Nation on Slate

2019 THI Scholar-in-Residence

Fighting the Future: The Philosophy of Predictive Algorithms in Criminal Justice

Thursday, April 11, 3:15 – 5:00 pm

Humanities 1 Bldg, Room 202

At different stages of the criminal justice system, from policing, bail hearings, and sentencing, computerized algorithms are replacing human decision-making in determining where to police, who to arrest, who goes to jail, and who goes free. This talk will introduce people to how these algorithms work, the under-appreciated moral problems with their implementation, and how the future of criminal justice depends on decisions we make now about the risks we are willing to tolerate for public safety.

Organized by The Humanities Institute and the Center for Public Philosophy

https://news.ucsc.edu/2019/04/philosophy-podcast-residency.html

All are welcome to attend the ML Poster Session next Thursday in the Engineering Courtyard.

Here is the poster describing the event.

All are welcome to a talk by PhD student Daniel Muthukrishna of Cambridge University 

Title:
Real-time classification of explosive transients using deep recurrent neural networks

Abstract:
We present RAPID (Real-time Automated Photometric IDentification), a novel time-series classification tool capable of automatically identifying transients from within a day of the initial alert, to the full lifetime of a light curve. Using a deep recurrent neural network with Gated Recurrent Units (GRUs), we present the first method specifically designed to provide early classifications of astronomical time-series data, typing 12 different transient classes. Our classifier can process light curves with any phase coverage, and it does not rely on deriving computationally expensive features from the data, making RAPID well-suited for processing the millions of alerts that ongoing and upcoming wide-field surveys such as the Zwicky Transient Facility (ZTF), and the Large Synoptic Survey Telescope (LSST) will produce. The classification accuracy improves over the lifetime of the transient as more photometric data becomes available, and across the 12 transient classes. We have begun running RAPID on the real-time ZTF survey, and have successfully classified several transients well before peak luminosity. We have made RAPID available as an open-source software package (https://astrorapid.readthedocs.io) for machine learning-based alert-brokers to use for the autonomous and quick classification of several thousand light curves within a few seconds.

Jeffrey Silverman, former NSF Astronomy postdoc and current Director of Data Science at Samba TV will present on: