£0 – £36

BMVA Technical Meeting: Cognitively inspired explainable perception-based A...

Event Information

Share this event

Date and Time

Location

Location

BCS (British Computer Society) in London.

5 Southampton St

London

WC2E 7HA

United Kingdom

View Map

Refund Policy

Refund Policy

Refunds up to 1 day before event

Event description

Description


BMVA Symposium: Cognitively inspired explainable perception-based AI

Organisers: Serge Thill, University of Plymouth, Maria Riveiro, University of Skövde,

Keynotes: Alessandra Sciutti, Italian Institute of Technology, Brad Hayes, University of Colorado Boulder & Yiannis Demiris, Imperial College,

Background:

AI systems are increasingly present in everyday society, from simple computer systems to agents such as autonomous vehicles or social robots. In this context, several researchers have noted that it is critical to understand how human users perceive such systems - in particular, the degree to which they understand how the system works, and what mental models they build of the underlying algorithms. "Explainable AI" thus refers to AI systems that behave or provide the necessary information so that their working becomes comprehensible to the human user. The need for transparency and explanations in AI-based systems to support interpretability, understandability, and trust has been highlighted recently by multiple authors from disparate disciplines in both AI and HCI communities and is, for example, the focus of Google's People & AI research initiative.

For this meeting, we are interested in AI systems that operate at least somewhat autonomously based on real-world sensory data (in particular, based on machine vision). This includes robotics and autonomous vehicles, but can also cover disembodied systems such as decision support systems.

We solicit contributions that address how such systems can produce explanations of their behaviour that are understood and accepted by human users and what effect (if any), these explanations have on factors such as trust, acceptance, and quality of interaction.

We are particularly interested in contributions that give detailed consideration to the fact that these are AI systems that sense (often through machine vision) the environment, and consider the possible role of understanding human cognitive mechanisms in the design of such systems. Relevant human cognitive mechanisms could include, for example, how humans perceive and interpret information themselves (which may be relevant in the design of explainable information processing by a machine) or how they interact with other intelligent agents (including their expectations on such interactions), which may impose constraints on the design of explainable systems that may be perceived as an intelligent, interactive agent by human users.

Important dates:

All those interested in presenting at this meeting are invited to submit a summary of their talk https://goo.gl/forms/OByON7vvlUX0xtDt1 by Dec 1, 2017

Queries

For queries please contact the organisers: serge.thill@plymouth.ac.uk, maria.riveiro@his.se

Share with friends

Date and Time

Location

BCS (British Computer Society) in London.

5 Southampton St

London

WC2E 7HA

United Kingdom

View Map

Refund Policy

Refunds up to 1 day before event

Save This Event

Event Saved