The Human Use of Machine Learning:
An Interdisciplinary Workshop

DECEMBER 16th, 2016 - VENICE

Any machine constructed for the purpose of making decisions, if it does not possess the power of learning, will be completely literal-minded.
Woe to us if we let it decide our conduct, unless we have previously examined its laws of action, and know fully that its conduct will be carried out on principles acceptable to us!

(N. Wiener, 1954)

Motivations

The development of machine learning has suggested ethical worries from its very beginning. Norbert Wiener, for example, as early as mid-fifties warned against a too simplistic idea of progress, divorced from any ethical guidance thereby reminding us the idea that there is no isolated “pure” science and that scientific and technological progress is in most of the cases socially and ethically mediated.
Inspired by that far-seeing reflection, we would like to stimulate a discussion starting from very basic questions such as: How do we reframe the notions of “wrong” and “good” in the context of machine learning and of data-driven technologies? What does it mean being responsible in today’s machine learning research and application? What is the place of moral values in the design process? How could we deal with engineering constrains, like efficacy or costs, in an ethical way? What is social good? What could be the social mission of machine learning? We do feel that this could provoke researchers to think critically about the philosophical underpinnings of today’s machine learning research encouraging a cross-fertilization of ideas and an “imaginative forward glance.”
With this workshop we aim at: 1) Stimulating potential collaborations among researchers with different background; 2) Identifying critical ethical questions and problems which may become the subject matter of future research projects within the SMC community and related areas; 3) Finding a common vocabulary to support the communication of technological achievements to non-expert people or to the large public; 4) Outlining how ethical aspects could be integrated into educational programme for future machine learning scientists.

Invited speakers

Bettina Berendt
University of Leuven
Krishna Gummadi
Max Planck Institute for Software Systems
Mireille Hildebrandt
Free University of Brussels
Arshak Navruzyan
Startup.ML

Fabio Roli
University of Cagliari
Viola Schiaffonati
Polytechnic University of Milan
Judith Simon
IT University of Copenhagen
Katherine Strandburg
New York University

Programme

Morning Session Afternoon Session
09:00 Welcome address [Slides] [Video] 14:10 Mireille Hildebrandt, The issue of bias: trade-offs and balance in ML [Abstract] [Slides] [Video]
09:10 Judith Simon, Big data & machine learning: Reflections on trust, trustworthiness & responsibility [Abstract] [Slides] [Video] 14:50 Krishna Gummadi, Discrimination in human vs. machine decision making [Abstract] [Slides] [Video]
09:50 Katherine Strandburg, Decision-making, machine learning and the value of explanation [Abstract] [Slides] [Video] 15:30 Michael Veale, How do public values get into public sector machine learning systems, if at all? [Abstract] [Slides][Video]
10:30 Coffee Break 15:45 Lucia Sommerer, Under the electronic eye: A legal evaluation of predictive policing [Abstract]
10:50 Fabio Roli, Some thoughts on safety of machine learning [Abstract] [Slides] [Video] 16:00 Coffee Break
11:30 Viola Schiaffonati, Preliminary steps for experimentally evaluating the impact of AI and robotic technologies [Abstract] [Slides] [Video] 16:20 Marília Monteiro, Innovating in democracy - machine learning and social participation: Can we hack law-making? [Abstract] [Slides] [Video]
12:10 Zackary C. Lipton, The mythos of model interpretability [Abstract] [Slides] [Video] 16:35 Matthias Treder, Mining the mind: Machine learning in brain research [Abstract] [Slides] [Video]
12:25 Federico Cabitza, The unintended consequences of chasing electric zebras [Abstract] [Slides] [Video] 16:50 Arshak Navruzyan, Avoiding bad machine learning predictions in critical decision domains [Abstract]
12:40 Lunch Break 17:30 Bettina Berendt, What does it mean to ask about "the human use of machine learning" - on positioning (not only) for engineers [Abstract] [Slides] [Video]
18:10 - 18:30 Open discussion and closing remarks

Organizers

Marcello Pelillo
Ca' Foscari University
Teresa Scantamburlo
Ca' Foscari University

Venue

The workshop will take place in the beautiful setting of the European Centre for Living Technology (ECLT), an interdisciplinary research centre devoted to study of complex systems and network science, located in the historical centre of Venice.

Contacts

European Centre for Living Technology, Ca' Minich, S. Marco 2940 30124 Venice

How to reach us
E-mail: eclt@unive.it
Website: http://www.ecltech.org/