Microlensing Hack Session

Webpage for the microlensing hack session at the CCA, January 30 -- February 1 2019.

Pages

Welcome

This hack session will focus on unsolved problems in modeling microlensing events. It will be held jointly with the annual microlensing workshop. The primary goal of the hack session is to engage people outside the traditional microlensing community in open problems in microlensing research.

Open problems include:

See below and the projects page for more detailed descriptions.

The resources page has a list of useful references including links to public tools and datasets.

Logistics

Microlensing 23 will be held January 28, 2019 – February 1, 2019 at the Center for Computational Astrophysics in New York City. A scientific conference will be held the first 2.5 days (Monday through Wednesday morning). The last 2.5 days (Wednesday afternoon through Friday) will be devoted to hacking. The draft SCHEDULE is available here.

Hack Session participants are welcome and encouraged to attend the full week to get a fully immersive introduction to microlensing. Many of the talks will cover recent analyses of microlensing data, including ongoing challenges, which will provide a sense of the state of the field. However, it is not necessary to attend the conference in order to participate in the hack session.

To register for the Conference and/or Hack Session, please visit the main conference website. ***Note registration for the conference is now closed. However, you can still register for the hack session.

What is the Purpose of the Hack Session? How Does it Work?

A hack session is an opportunity for focused work on a particular problem. The goal is to foster novel solutions to problems by bringing together people with different areas of expertise. In this case, the goal is to bring together microlensing experts and their knowledge of modeling and analysis problems in the field with experts in machine learning, statistics, and mathematics and their knowledge of the latest techniques for large-scale time-series data analysis, methods for likelihood maximization, and strategies for computational challenges related to finding the roots of high-order complex polynomials.

A hack session is meant to be freeform to allow collaborations to develop naturally. However, we anticipate two types of work will tend to dominate:

  1. Open discussions of particular unsolved problems.

To facilitate this, we will be organizing breakout sessions on Thursday and Friday based on topics brought by participants. These discussions will be about an hour in length, but may lead to further collaboration. An appropriate length for introducing a topic would be 5 minutes or less.

  1. Focused work (e.g., coding) on a specific project that can be completed in a two-day timeframe.

For example, at one-day AAS hack sessions, participants have completed a range of projects from writing a Twitterbot to tweet flight information for the local airport to writing new modules for the Python package astropy. In this case, the projects of most interest will provide opportunities for machine learning or data analysis experts to apply their expertise.

It is expected that microlensing participants will bring projects or topics they would like to work on. At the same time, they should also be open to allowing collaborations to take them in new directions.

For those not familiar with microlensing, this field deals with the detection of planets in time-series photometric data. It is similar to Kepler transiting planet searches in the large number of datapoints (observational cadence of 1/15 min to 1/day for hundreds of millions of stars) and small number of true positives (dozens per year from the ground or hundreds from space). It differs from Kepler in that the signals, modeling, and likelihood surfaces are much more complex.

Guidelines for Collaboration

To ensure transparency and openness, we have adopted the collaboration policy developed for the Gaia Sprints. This policy recommends that participants agree to the following:

All participants will be expected to openly share their ideas, expertise, code, and interim results. Project development will proceed out in the open, among participants and in the world.

Participants will be encouraged to change gears, start new collaborations, and combine projects. Any participant who contributes significantly to a project can expect co-authorship on resulting scientific papers, and any participant who gets significant contributions to a project is expected to include those contributors as co-authors.

These rules make it inadvisable to bring proprietary data sets or proprietary code, unless the participant bringing such assets has the rights to open them or add collaborators.

Be a Nice Person

Code of Conduct

The organizers are committed to making this meeting productive and enjoyable for everyone, regardless of gender, sexual orientation, disability, physical appearance, body size, race, nationality or religion. We will not tolerate harassment of participants in any form.

Please follow these guidelines:

Participants asked to stop any inappropriate behaviour are expected to comply immediately. Attendees violating these rules may be asked to leave the event at the sole discretion of the organizers without a refund of any charge.

Any participant who wishes to report a violation of this policy is asked to speak, in confidence, to Calen Henderson chenderson@ipac.caltech.edu and/or Rachel Street rstreet@lco.global.

The contact people are volunteers without formal training who can’t give legal advice. Their role is to provide support and to facilitate concerns being addressed. Anyone with concerns is encouraged to speak to one of the contact people and may do so confidentially to request advice, which will be their personal interpretation and guidance. Both the person reporting a concern and the contact person should write down a full description of their concern or incident as soon as possible after the event. The contact person will not share that information with anyone without the permission of the person reporting a concern. The contact person may, if the reporting person is willing and the contact person feels comfortable doing do, discuss the issue with someone named in the complaint and/or the other contact person. If so, these conversations should also be documented. If the reporting person wishes to make a formal complaint, they or the contact person will contact codeofconduct@simonsfoundation.org, and provide the documentation described above.

This code of conduct is based on the “London Code of Conduct”, as originally designed for the conference “Accurate Astrophysics. Correct Cosmology”, held in London in July 2015.

Scientific Organizing Committee

Jennifer Yee (Chair)
Etienne Bachelet
David Bennett
Dan Foreman-Mackey
Calen Henderson
David Hogg
Rodrigo Luger
Radek Poleski
Clement Ranc
Yossi Shvartzvald
Rachel Street