Computational Interaction

Abstract

Computational design is aimed at supporting or automating design processes using computational techniques. However, some classes of design tasks involve criteria that are di cult to handle only with computers. For example, visual design tasks seeking to ful ll aesthetic goals are di cult to handle purely with computers. One promising approach is to leverage human computation; that is, to incorporate human input into the computation process. Crowdsourcing platforms provide a convenient way to integrate such human computation into a working system.

Adapting User Interfaces with Model-based Reinforcement Learning

“…The prevailing understanding is that learning is a key capability for adaptive systems [32]. Two learning capabilities are needed: (1) inference, the capability to update assumptions about the user based on observations; and (2) decision-making, the capability to choose appropriate adaptation in the light of assumptions about the user [41]. The two challenges can be relaxed, for example if user state is trivially known, or if the state is highly predictive of appropriate adaptation.…”

A Probabilistic Interpretation of Motion Correlation Selection Techniques

“…In this work, we take direct inspiration from the pioneering work of Williamson and Murray-Smith, abstracting it to model later approaches and extending it to enable the analysis and design of new interfaces. Our work is framed within the larger program of computational interaction, which "focuses on the use of algorithms and mathematical models to explain and enhance interaction" [23]. Core to computational interaction is mathematical modelling, typically involving ways for updating the model using data observed from the user [23].…”

“…Our work is framed within the larger program of computational interaction, which "focuses on the use of algorithms and mathematical models to explain and enhance interaction" [23]. Core to computational interaction is mathematical modelling, typically involving ways for updating the model using data observed from the user [23]. In this paper, we model motion correlation using a probabilistic approach, which enables us to derive a set of techniques for examining motion correlation designs before collecting user data.…”

“…In this paper, we propose an interpretation of motion correlation based on probability and information theory, inspired by the early work conducted by Williamson and Murray-Smith [34,35] and aligned with current trends in computational interaction [23]. As such, we offer the following contributions: (1) We demonstrate that previous works on motion correlation can be modelled as a probabilistic reasoning problem; (2) we demonstrate a series of probabilistic techniques for better understanding combinations of interface design decisions; (3) we offer a case study of the application of these techniques in the context of a motion correlation technique, Orbits [10].…”

Using Simulation to Aid the Design and Optimization of Intelligent User Interfaces for Quality Assurance Processes in Machine Learning

“…It is desirable for the design of a QA4ML UI to optimize (a) according to (b), (c), (d), etc. This problem falls into the topic of computational interaction [25], especially interface and interaction modeling [1, 3-5, 12-14, 22, 24, 34], and is related to the evaluation method of action analysis [5,7,16,17,27].…”