Life at the academic coalface: validation of a holistic academic workload estimation tool

Kenny, John; Fluck, A

doi:10.1007/s10734-022-00912-x

Cited by 9 publications

(4 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, emphasis is still placed on the importance of conciseness, and it would be rare for a thesis to exceed 300 pages. Given the ever-increasing academic workloads [5,6], this can be appreciated by doctoral examiners who examine the texts.…”

Section: Traditional Thesismentioning

confidence: 99%

Supporting Doctoral Candidates through Completion and Final Examination

Riches-Suman

2024

Encyclopedia

View full text Add to dashboard Cite

Completion and final examination comprise the final stages of a doctoral program and represent the culmination of the doctoral candidates’ years of research. In this entry, completion is defined as the writing and submission of a doctoral thesis, and final examination is defined as the viva voce. Over the years, the format and scope of doctoral degrees has expanded and a variety of formats are now offered. In addition to the traditional research-only doctoral degree, professional, practice-based, and new route programs also contain a taught element alongside research. However, the creation of a substantive thesis or practice-based alternative addressing a novel research question is common to all. In contrast, processes and formats of viva voces vary across the globe. These range from private, closed-door defenses to assessed or ritualistic public defense presentations. For both completion and final examination, there are many practical and psychological hurdles that need to be navigated in order for the candidate to attain their doctoral degree. This entry will highlight these aspects as well as provide evidence-based guidance for supervisors in supporting their doctoral candidates through these daunting final stages.

show abstract

Section: Traditional Thesismentioning

confidence: 99%

Supporting Doctoral Candidates through Completion and Final Examination

Riches-Suman

2024

Encyclopedia

View full text Add to dashboard Cite

show abstract

“…To ensure task-informativeness, we implement hierarchical feature extractors that learn task-shared and task-specific feature representations for tasks with different learning difficulties. Inspired by human learning 57,58 and the work of Guo et al, 59 we separate the total of M tasks into M 1 easy-to-learn and M 2 hard-to-learn tasks based on the task difficulty, which is inversely proportional to the learning performance (e.g., prediction accuracy) in the source domain. In particular, we train, respectively, M classifiers with the same model complexity using source domain data of the M tasks, and obtain testing accuracy values, { p 1 , p 2 , …, p M }, for the M tasks.…”

Section: Hiermud: a Hierarchical Multi-task Unsupervised Domain Adapt...mentioning

confidence: 99%

HierMUD: Hierarchical multi-task unsupervised domain adaptation between bridges for drive-by damage diagnosis

Liu

Bergés

et al. 2022

Structural Health Monitoring

View full text Add to dashboard Cite

Monitoring bridges through vibration responses of drive-by vehicles enables efficient and low-cost bridge maintenance by allowing each vehicle to inspect multiple bridges and eliminating the needs for installing and maintaining sensors on every bridge. However, many existing drive-by monitoring approaches are based on supervised learning models that require massive labeled data from every bridge. It is expensive and time-consuming, if not impossible, to obtain these labeled data. Furthermore, directly applying a supervised learning model trained on one bridge to new bridges would result in low accuracy due to the shift between different bridges’ data distributions. Moreover, when we have multiple tasks (e.g., damage detection, localization, and quantification), the distribution shifts become more challenging than having only one task because different tasks have distinct distribution shifts and varying task difficulties. To this end, we introduce HierMUD, the first Hierarchical Multi-task Unsupervised Domain adaptation framework that transfers the damage diagnosis model learned from one bridge to a new bridge without requiring any labels from the new bridge. Specifically, our framework learns a hierarchical neural network model in an adversarial way to extract features that are informative to multiple tasks and invariant across multiple bridges. To match distributions over multiple tasks, we design a new loss function based on a newly derived generalization risk bound to adaptively assign higher weights to tasks with more shifted distributions. To learn multiple tasks with varying task difficulties, we split them into easy-to-learn and hard-to-learn tasks based on their distributions. Then, we formulate a feature hierarchy to utilize more learning resources to improve the hard-to-learn tasks’ performance. We evaluate our framework with experimental data from 2 bridges and 3 vehicles. We achieve up to 2X better performance than baseline methods, including average accuracy of 95% for damage detection, 93% for localization, and 0.38 lbs mean absolute error for quantification.

show abstract

“…In a case study for the development of a quantifiable academic workload model, Kenny et al state that a lack of clarity about how to quantify academic workload leads to deterioration of academic performance [6]. They also stated that the nature and extent of academic work must be accounted for credibly and transparently and that the allocation of workload should account for these factors.…”

Section: Quantifying Course Workloadmentioning

confidence: 99%

“…Kenny et al also notes that defining realistic time limits is a necessary step towards ensuring welfare of academic staff [2]. One of the big reasons for this was identified as budgetary concerns, with universities being unwilling to hire additional faculty to address the workload inequity [6].…”

Section: Workload Inequity In Teaching Allocationmentioning

confidence: 99%

Holistic teaching workload allocation for research-intensive universities

Roopchandani

View full text Add to dashboard Cite

The above methods and techniques were implemented and validated using real-world data, and the results were found to be highly effective, accurately quantifying the workload in a variety of scenarios, and allocating lectures, tutorials, and labs to faculty members while minimizing fragmentation and uneven faculty overloading. The entire teaching workload allocation process involving 1152 lectures, tutorials, and labs took only 30 seconds on a standard PC. Finally, the proposed techniques ensure an inclusive approach by incorporating the preferences of students, faculty, and management while ensuring an equitable proportion of lectures, tutorials, and labs for each faculty member. vi

show abstract

Life at the academic coalface: validation of a holistic academic workload estimation tool

Cited by 9 publications

References 30 publications

Supporting Doctoral Candidates through Completion and Final Examination

Supporting Doctoral Candidates through Completion and Final Examination

HierMUD: Hierarchical multi-task unsupervised domain adaptation between bridges for drive-by damage diagnosis

Holistic teaching workload allocation for research-intensive universities

Contact Info

Product

Resources

About