Designing for effective end-user interaction with machine learning

Amershi, Saleema

doi:10.1145/2046396.2046416

Cited by 44 publications

(34 citation statements)

References 60 publications

(78 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, an agent should show steady behaviour by default, but nevertheless apply reward immediately and appropriately [41,7]. Additionally, domain experts should be provided with a variety of complex feedback and control mechanisms [43], because users favour transparency and are willing to learn how a system works to give nuanced feedback [44]. Even though, the user should be supported with an appropriate level of guidance and offered summaries and explanations of system behaviour, which should preferably be lightweight but also scalable [43].…”

Section: Essential Aspects Of Interactive Reinforcement Learningmentioning

confidence: 99%

“…Additionally, domain experts should be provided with a variety of complex feedback and control mechanisms [43], because users favour transparency and are willing to learn how a system works to give nuanced feedback [44]. Even though, the user should be supported with an appropriate level of guidance and offered summaries and explanations of system behaviour, which should preferably be lightweight but also scalable [43]. For example, a visualization of higher level features of convolutional layers in deep networks [39], as was used in [21].…”

Section: Essential Aspects Of Interactive Reinforcement Learningmentioning

confidence: 99%

See 1 more Smart Citation

Applying and Augmenting Deep Reinforcement Learning in Serious Games through Interaction

Dobrovsky

Borghoff

Hofmann

2017

Period. Polytech. Elec. Eng. Comp. Sci.

View full text Add to dashboard Cite

IntroductionThe term serious game (SG) origins from [1] and refers to games not exclusively developed for mere entertainment, but primarily for creating educational value. SGs are counted among the current e-learning trends and are gaining more and more acceptance and influence [2]. SG development slightly differs from that of entertainment games: SGs are often individual products for a restricted target audience, industrial branch or company. This results in high expenditure and deficient reusability, although the market shows growing interest in cost-efficient and customized applications [2]. One of the most relevant, but also most time-consuming tasks is the creation of reasonable, human-like non-player character (NPC) behaviour [3][4][5]. Thus, simplifying authoring and adaptation of AI in SGs seems desirable and profitable.Machine learning, especially reinforcement learning (RL), is occasionally used for automated NPC behaviour generation. Nevertheless, several issues arise when applying RL to create believable and diverse behaviour in complex scenarios. We indicate a way to overcome these issues by combining RL with human guidance, including effective collaboration between a learning system and human experts.In this paper, we give a short introduction to some challenges of behaviour generation in serious games and to the background of RL, deep reinforcement learning (DRL) and interactive reinforcement learning (iRL). We show related approaches and depict current issues of applying DRL methods to SGs. Furthermore, we introduce SanTrain as a SG providing challenging scenarios for NPC behaviour generation. Finally, we show how our approach of interactive deep reinforcement learning (iDRL), integrated into a flexible framework, could enhance AI development in SGs and exemplarily indicate valuable application opportunities in SanTrain.

show abstract

Section: Essential Aspects Of Interactive Reinforcement Learningmentioning

confidence: 99%

Section: Essential Aspects Of Interactive Reinforcement Learningmentioning

confidence: 99%

Applying and Augmenting Deep Reinforcement Learning in Serious Games through Interaction

Dobrovsky

Borghoff

Hofmann

2017

Period. Polytech. Elec. Eng. Comp. Sci.

View full text Add to dashboard Cite

show abstract

“…As an alternative, researchers have investigated mixedinitiat ive interfaces [15] that keep humans in the loop, provid ing proper feedback and domain knowledge to machine learning algorithms [1], [2], [7], [12], [18], [21], [26]. For examp le, CueFlik [12] allows end-users to locate images on the web through a combination of keywo rd search and iterative examp le-based concept refinement activities.…”

Section: Interactive Machine Learningmentioning

confidence: 99%

“…The choice "Very impo rtant" corresponds to a weight of 1 (highest), "not important" corresponds to a weight of 0 (lo west), the weights of "important" features are set to be in a range of [0.6, 1] wh ile "not sure" features are set to be within [0,1]. In the end, we get a set of ranges for the weights of all features:…”

Section: Constraint Conflict Detectionmentioning

confidence: 99%

MindMiner: A Mixed-Initiative Interface for Interactive Distance Metric Learning

Fan

Liu

Cao

et al. 2015

Human-Computer Interaction – INTERACT 2015

View full text Add to dashboard Cite

“…A number of systems have explored using active learning for assisting users in defining personalized concepts [1,9,21]. For example, [9] supports users in interactively defining concepts for re-ranking web images search results, and presents users the option of classifying images that are closest to the decision boundary between positive and negative classes, using a nearest-neighbor approach.…”

Section: Active Learningmentioning

confidence: 99%

Active learning of intuitive control knobs for synthesizers using gaussian processes

Huang

Duvenaud

Arnold

et al. 2014

Proceedings of the 19th International Conference on Intelligent User Interfaces

View full text Add to dashboard Cite

Typical synthesizers only provide controls to the low-level parameters of sound-synthesis, such as wave-shapes or filter envelopes. In contrast, composers often want to adjust and express higher-level qualities, such as how 'scary' or 'steady' sounds are perceived to be.We develop a system which allows users to directly control abstract, high-level qualities of sounds. To do this, our system learns functions that map from synthesizer control settings to perceived levels of high-level qualities. Given these functions, our system can generate high-level knobs that directly adjust sounds to have more or less of those qualities. We model the functions mapping from control-parameters to the degree of each high-level quality using Gaussian processes, a nonparametric Bayesian model. These models can adjust to the complexity of the function being learned, account for nonlinear interaction between control-parameters, and allow us to characterize the uncertainty about the functions being learned.By tracking uncertainty about the functions being learned, we can use active learning to quickly calibrate the tool, by querying the user about the sounds the system expects to most improve its performance. We show through simulations that this model-based active learning approach learns high-level knobs on certain classes of target concepts faster than several baselines, and give examples of the resulting automaticallyconstructed knobs which adjust levels of non-linear, highlevel concepts.

show abstract

Designing for effective end-user interaction with machine learning

Cited by 44 publications

References 60 publications

Applying and Augmenting Deep Reinforcement Learning in Serious Games through Interaction

Applying and Augmenting Deep Reinforcement Learning in Serious Games through Interaction

MindMiner: A Mixed-Initiative Interface for Interactive Distance Metric Learning

Active learning of intuitive control knobs for synthesizers using gaussian processes

Contact Info

Product

Resources

About