ADS Coffee & Data offers the opportunity for researchers and business to share their knowledge and give insight on a central theme, specifically on Friday 12 May this will be on Artificial Intelligence and more specifically on learning from user provided data. There is also a chance to network with a cup of coffee.
Introduction & Chair: Maarten de Rijke (Informatics Institute, UvA) (https://staff.fnwi.uva.nl/m.derijke/)
09:15-09:35: Julia Kiseleva (Informatics Institute, UvA and UserSat (http://juliakiseleva.com)) on Evaluating Personal Assistants
There is a rapid growth in the use of voice-controlled intelligent personal assistants on mobile devices, such as Microsoft’s Cortana, Google Now, and Apple’s Siri. They significantly change the way users interact with search systems, not only because of the voice control use and touch gestures, but also due to the dialogue-style nature of the interactions and their ability to preserve context across different queries. In this talk, we propose an automatic method to predict user satisfaction with intelligent assistants that exploits all the interaction signals, including voice commands and physical touch gestures on the device. First, we conduct an extensive user study to measure user satisfaction with intelligent assistants, and simultaneously record all user interactions. Second, we show that the dialogue style of interaction makes it necessary to evaluate the user experience at the overall task level as opposed to the query level. Third, we train a model to predict user satisfaction, and find that interaction signals that capture the user reading patterns have a high impact.
09:35-09:55: Zeynep Akata (Informatics Institute, UvA) (https://www.mpi-inf.mpg.de/departments/computer-vision-and-multimodal-computing/people/zeynep-akata/)on Vision and Language for Multimodal Deep Learning: Towards Explanatory AI
Scaling up visual category recognition to large numbers of classes remains challenging. A promising research direction is zero-shot learning, which does not require any labeled training data to recognize new classes, but rather relies on some form of auxiliary information describing the new classes. Ultimately, this may allow to use textbook knowledge that humans employ to learn about new classes by transferring knowledge from classes they know well. We tackle the zero-shot learning problem by learning a compatibility function such that matching image-class embedding pairs are assigned a higher score than mismatching pairs; zero-shot classification proceeds by finding the label vector yielding the highest joint compatibility score. We use detailed visual descriptions collected from naive users as side-information for zero-shot learning and also to translate characters to image pixels. Moreover, we generate visual explanations which justify a classification decision through fine-grained image captioning.
09:55-10:35: Katja Hofmann (Microsoft Research Cambridge (https://www.microsoft.com/en-us/research/people/kahofman/)) on Interactive Machine Learning: from Bandits to Collaborative AI
A long-standing goal of artificial intelligence (AI) research is to develop artificial agents that can assist people in tasks that are tedious, strenuous, or dangerous. Achieving this goal would require AI to deal with the complex and constantly changing real-world environments we live in. Interactive machine learning encompasses techniques that can address the challenge of learning in constantly changing, complex environments. In this talk I introduce some of our recent work in this area, starting with contextual bandit algorithms that can interactively learn to assist with restaurant recommendation, to first steps towards developing collaborative AI in the Minecraft world.
10:35 – 11:00 Coffee
Registration is free but please do so in advance through Meet-up
The event will be in English and is open to all
Amsterdam Data Science (ADS) accelerates data science research by connecting, sharing and showcasing world-class technology, expertise and talent from Amsterdam on a regional, national and international level. Our research enables business and society to better gather, store, analyse and present data in order to gain valuable insights and make informed decisions.
Find out more about ADS at http://amsterdamdatascience.nl