Optimizing User Interface Layouts via Gradient Descent

Duan, Peitong; Wierzynski, Casimir; Nachman, Lama

doi:10.1145/3313831.3376589

Cited by 19 publications

(16 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MobilenetV2 is preferred as it has higher accuracy and a model size of less than 10MB. Figure 7 shows that similar existing solutions [8,9] do not handle aesthetics and saliency. It can be seen in Fig.…”

Section: Resultsmentioning

confidence: 99%

“…Recent attempts in this field have used Machine-learning algorithms for UI layout optimization. Duan et al [8] used gradient descent algorithm to train a model to optimize layouts for task performance by generating better alternatives of the layouts. The authors recognize that human intervention is required to refine the UI for improving its aesthetics.…”

Section: Ai In User Interfacesmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic User Interface Composition

Kumar¹,

Natarajan²,

Shariff³

et al. 2021

Communications in Computer and Information Science

View full text Add to dashboard Cite

With continuous innovation and a persistent attempt to achieve natural human machine interactions, user interfaces have evolved into the current phase of Voice User Interfaces. Voice User Interfaces augmented with visual information are becoming prominent in a variety of devices with different form-factors. Developing user interfaces for such a wide range of display configurations is a challenging task. This paper puts forward an approach to dynamically compositing such user interfaces without compromising on User Experience. This work examines details of devising a neural network model to automate user interface layout composition based on aesthetics and saliency. A 40000 dataset was created for this work with eight trained annotators. Ground truth is estimated from the above annotations using Expectation Maximization algorithm. Experiments show that users are highly satisfied with the model results. This model is deployed in Lock Screen auto layout available in latest Samsung Android phones.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Ai In User Interfacesmentioning

confidence: 99%

Dynamic User Interface Composition

Kumar¹,

Natarajan²,

Shariff³

et al. 2021

Communications in Computer and Information Science

View full text Add to dashboard Cite

show abstract

“…Soundr [ 172 ] used Deep Learning to figure out the user’s spatial location and head orientation using voice. Duan et al [ 173 ] developed a technique to optimize UI interfaces automatically with error correction. Pfau et al [ 174 ] looked at how Deep Learning techniques can improve dynamic difficulty adjustments in games.…”

Section: Classifying Hcml Researchmentioning

confidence: 99%

“…Considering the works that focused on the user side, some researchers catered to general end-users or consumers [83,101,200,210], while others on specific end-users. Examples for these include people who need assistance [2,80,[86][87][88][89]96,117,147], medical professionals [57,67,110,192,193], international travelers [50], Amazon Mechanical Turk [60,99], drivers [161,162], musicians [102], teachers [124], students [128], children [72,125], UX designers [65,115,206,209], UI designers [103,111,173], data analysts [97], video creators [84], and game designers [70,165,174,211]. Apart from focusing on a specific user group, some have tried to understand multiple user-perspectives from ML engineers to the end-user [48].…”

Section: The 'Human' In Hcmlmentioning

confidence: 99%

A Review of Recent Deep Learning Approaches in Human-Centered Machine Learning

Kaluarachchi

Reis

Nanayakkara

2021

Sensors

View full text Add to dashboard Cite

After Deep Learning (DL) regained popularity recently, the Artificial Intelligence (AI) or Machine Learning (ML) field is undergoing rapid growth concerning research and real-world application development. Deep Learning has generated complexities in algorithms, and researchers and users have raised concerns regarding the usability and adoptability of Deep Learning systems. These concerns, coupled with the increasing human-AI interactions, have created the emerging field that is Human-Centered Machine Learning (HCML). We present this review paper as an overview and analysis of existing work in HCML related to DL. Firstly, we collaborated with field domain experts to develop a working definition for HCML. Secondly, through a systematic literature review, we analyze and classify 162 publications that fall within HCML. Our classification is based on aspects including contribution type, application area, and focused human categories. Finally, we analyze the topology of the HCML landscape by identifying research gaps, highlighting conflicting interpretations, addressing current challenges, and presenting future HCML research opportunities.

show abstract

“…in ability-based user interfaces [33,34]. More recent work trained a neural network to predict users' task performance from a previously collected data [28]. Closed-loop adaptive systems, i.e.…”

Section: Adaptive Design Through Optimization In Hcimentioning

confidence: 99%