Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Personalizing Mixed Reality UIs from User Interactions
Multi-objective discrete optimization schemes have demonstrated their effectiveness in automatically generating UI layouts for virtual and augmented reality desks [1,2]. However, the quality of the solutions heavily relies on the weighting assigned to different objectives. In this thesis, our objective is to discover an optimal combination of weights that accurately captures the requirements and preferences of an individual user during a single session. To achieve this, we propose incorporating user adjustments of an initially generated UI as feedback to inform the weight updates. By doing so, our aim is to personalize mixed-reality desks with minimal user input, streamlining the customization process.
Keywords: Discrete optimization
Bayesian optimization
Probabilistic user modeling
**Literature**
[1] Lindlbauer, David, Anna Maria Feit, and Otmar Hilliges. "Context-aware online adaptation of mixed reality interfaces." Proceedings of the 32nd annual ACM symposium on user interface software and technology. 2019
[2] Cheng, Yifei, et al. "Semanticadapt: Optimization-based adaptation of mixed reality layouts leveraging virtual-physical semantic connections." The 34th Annual ACM Symposium on User Interface Software and Technology. 2021.
**Required skills**
- Solid programming skills (C#, Python, or similar), experience with Unity is a plus
- Prior experience with optimization, knowledge in Bayesian approaches are a plus
- Self-motivated with strong investigative attitude, able to research and work alone
**Literature**
[1] Lindlbauer, David, Anna Maria Feit, and Otmar Hilliges. "Context-aware online adaptation of mixed reality interfaces." Proceedings of the 32nd annual ACM symposium on user interface software and technology. 2019
[2] Cheng, Yifei, et al. "Semanticadapt: Optimization-based adaptation of mixed reality layouts leveraging virtual-physical semantic connections." The 34th Annual ACM Symposium on User Interface Software and Technology. 2021.
**Required skills**
- Solid programming skills (C#, Python, or similar), experience with Unity is a plus
- Prior experience with optimization, knowledge in Bayesian approaches are a plus
- Self-motivated with strong investigative attitude, able to research and work alone
Extend existing optimization algorithms to personalize UIs from few user interactions
Extend existing optimization algorithms to personalize UIs from few user interactions
Christoph Gebhardt (christoph.gebhardt@inf.ethz.ch)
Christoph Gebhardt (christoph.gebhardt@inf.ethz.ch)