Open Access Theses and Dissertations

Advanced search options

Advanced Search Options 🞨

Browse by author name (“Author name starts with…”).

Find ETDs with:

in
/  
in
/  
in
/  
in

Written in Published in Earliest date Latest date

Sorted by

Results per page:

Sorted by: relevance · author · university · date | New search

You searched for +publisher:"Delft University of Technology" +contributor:("Wang, Ziqi"). Showing records 1 – 3 of 3 total matches.

Search Limiters

Last 2 Years | English Only

No search limiters apply to these results.

▼ Search Limiters


Delft University of Technology

1. Anand, Kanav (author). Black Magic in Deep Learning: Understanding the role of humans in hyperparameter optimization.

Degree: 2019, Delft University of Technology

URL: http://resolver.tudelft.nl/uuid:7a9df9fb-5dc4-4d72-a966-45edbb2bc942 

Deep learning is proving to be a useful tool in solving problems from various domains. Despite a rich research activity leading to numerous interesting deep learning models, recent large scale studies have shown that with hyperparameter optimization it is hard to distinguish these models based on their final performance. Hyperparameter optimization has shown to improve the state of the art results on several occasions. These results cast the doubts over the performance of these improved deep learning models and lead to the question whether the final performance of a deep learning model is dependent on the person performing the hyperparameter optimization task. A user study was conducted to evaluate the impact of human's prior experience in deep learning on the final performance of a deep learning model. 31 people with different levels of experience in deep learning were invited to perform a hyperparameter optimization task. The collected data was analyzed to find the relationship between human and the final performance of the deep learning model used for the user study. From the results, we observed that the final performance of the model vary with every participant, and a strong correlation between the participant's experience and the final performance achieved. Our data suggest that an experienced participant finds better results using fewer resources.

Computer Science

Advisors/Committee Members: van Gemert, Jan (mentor), Loog, Marco (graduation committee), Wang, Ziqi (mentor), Delft University of Technology (degree granting institution).

Subjects/Keywords: hyperparameter optimization; deep learning; machine learning; user study

Record Details  Similar Records    Google PlusoneFacebookTwitterCiteULikeMendeleyreddit

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Anand, K. (. (2019). Black Magic in Deep Learning: Understanding the role of humans in hyperparameter optimization. (Masters Thesis). Delft University of Technology. Retrieved from http://resolver.tudelft.nl/uuid:7a9df9fb-5dc4-4d72-a966-45edbb2bc942

Chicago Manual of Style (16th Edition):

Anand, Kanav (author). “Black Magic in Deep Learning: Understanding the role of humans in hyperparameter optimization.” 2019. Masters Thesis, Delft University of Technology. Accessed February 28, 2021. http://resolver.tudelft.nl/uuid:7a9df9fb-5dc4-4d72-a966-45edbb2bc942.

MLA Handbook (7th Edition):

Anand, Kanav (author). “Black Magic in Deep Learning: Understanding the role of humans in hyperparameter optimization.” 2019. Web. 28 Feb 2021.

Vancouver:

Anand K(. Black Magic in Deep Learning: Understanding the role of humans in hyperparameter optimization. [Internet] [Masters thesis]. Delft University of Technology; 2019. [cited 2021 Feb 28]. Available from: http://resolver.tudelft.nl/uuid:7a9df9fb-5dc4-4d72-a966-45edbb2bc942.

Council of Science Editors:

Anand K(. Black Magic in Deep Learning: Understanding the role of humans in hyperparameter optimization. [Masters Thesis]. Delft University of Technology; 2019. Available from: http://resolver.tudelft.nl/uuid:7a9df9fb-5dc4-4d72-a966-45edbb2bc942


Delft University of Technology

2. Yin, Z. (author). Assessment of Parkinson's Disease Severity from Videos using Deep Architectures.

Degree: 2020, Delft University of Technology

URL: http://resolver.tudelft.nl/uuid:a0336a50-d169-45cb-abe7-097ba8d15084 

Parkinson's disease (PD) diagnosis is based on clinical criteria, i.e. bradykinesia, rest tremor, rigidity, etc. Assessment of the severity of PD symptoms, however, is subject to inter-rater variability. In this paper, we propose a deep learning based automatic PD diagnosis method using videos recorded during the assessment with the Movement Disorders Society - Unified PD rating scale (MDS-UPDRS) part III. Seven tasks from the MDS-UPDRS III are investigated, which show the symptoms of bradykinesia and postural tremors. We demonstrate the effectiveness of automatic classification of PD severity using 3D Convolutional Neural Network (CNN) and the PD severity classification can benefit from non-medical datasets for transfer learning. We further design a temporal self-attention (TSA) model to focus on the subtle temporal vision changes in our PD video dataset. The temporal relative self-attention-based 3D CNN classifier gives promising classification results on task-level videos. We also propose a task-assembling method to predict the patient-level severity through stacking classifiers. We show the effectiveness of TSA and task-assembling method on our PD video dataset empirically. Advisors/Committee Members: van Gemert, J.C. (mentor), Dibeklioglu, Hamdi (mentor), Wang, Huijuan (graduation committee), Wang, Ziqi (mentor), Geraedts, Victor (graduation committee), Delft University of Technology (degree granting institution).

Subjects/Keywords: Parkinson's Disease; Deep learning; Transfer learning; Self-attention; Multi-domain learning

Record Details  Similar Records    Google PlusoneFacebookTwitterCiteULikeMendeleyreddit

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Yin, Z. (. (2020). Assessment of Parkinson's Disease Severity from Videos using Deep Architectures. (Masters Thesis). Delft University of Technology. Retrieved from http://resolver.tudelft.nl/uuid:a0336a50-d169-45cb-abe7-097ba8d15084

Chicago Manual of Style (16th Edition):

Yin, Z (author). “Assessment of Parkinson's Disease Severity from Videos using Deep Architectures.” 2020. Masters Thesis, Delft University of Technology. Accessed February 28, 2021. http://resolver.tudelft.nl/uuid:a0336a50-d169-45cb-abe7-097ba8d15084.

MLA Handbook (7th Edition):

Yin, Z (author). “Assessment of Parkinson's Disease Severity from Videos using Deep Architectures.” 2020. Web. 28 Feb 2021.

Vancouver:

Yin Z(. Assessment of Parkinson's Disease Severity from Videos using Deep Architectures. [Internet] [Masters thesis]. Delft University of Technology; 2020. [cited 2021 Feb 28]. Available from: http://resolver.tudelft.nl/uuid:a0336a50-d169-45cb-abe7-097ba8d15084.

Council of Science Editors:

Yin Z(. Assessment of Parkinson's Disease Severity from Videos using Deep Architectures. [Masters Thesis]. Delft University of Technology; 2020. Available from: http://resolver.tudelft.nl/uuid:a0336a50-d169-45cb-abe7-097ba8d15084


Delft University of Technology

3. Li, Jiahui (author). Attention-Aware Age-Agnostic Visual Place Recognition.

Degree: 2019, Delft University of Technology

URL: http://resolver.tudelft.nl/uuid:250d37a9-bc0d-4f8f-8d1a-d31a98dc22d7 

A cross-domain visual place recognition (VPR) task is proposed in this work, i.e., matching images of the same architectures depicted in different domains. VPR is commonly treated as an image retrieval task, where a query image from an unknown location is matched with relevant instances from geo-tagged gallery database. Different from conventional VPR settings where the query images and gallery images come from the same domain, we propose a more common but challenging setup where the query images are collected under a new unseen condition. The two domains involved in this work are contemporary street view images of Amsterdam from the Mapillary dataset (source domain) and historical images of the same city from Beeldbank dataset (target domain). We tailored an age-invariant feature learning CNN that can focus on domain invariant objects and learn to match images based on a weakly supervised ranking loss. We propose an attention aggregation module that is robust to domain discrepancy between the train and the test data. Further, a multi-kernel maximum mean discrepancy (MK-MMD) domain adaptation loss is adopted to improve the cross-domain ranking performance. Both attention and adaptation modules are unsupervised while the ranking loss uses weak supervision. Visual inspection shows that the attention module focuses on built forms while the dramatically changing environment are less weighed. Our proposed CNN achieves state of the art results (99% accuracy) on the single-domain VPR task and 20% accuracy at its best on the cross-domain VPR task, revealing the difficulty of age-invariant VPR. Advisors/Committee Members: van Gemert, Jan (mentor), Khademi, Seyran (mentor), Wang, Ziqi (mentor), Reinders, Marcel (graduation committee), Nan, Liangliang (graduation committee), Delft University of Technology (degree granting institution).

Subjects/Keywords: Computer Vision; Domain Adaptation; Image Matching; Attention Mechanism

Record Details  Similar Records    Google PlusoneFacebookTwitterCiteULikeMendeleyreddit

❌

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Li, J. (. (2019). Attention-Aware Age-Agnostic Visual Place Recognition. (Masters Thesis). Delft University of Technology. Retrieved from http://resolver.tudelft.nl/uuid:250d37a9-bc0d-4f8f-8d1a-d31a98dc22d7

Chicago Manual of Style (16th Edition):

Li, Jiahui (author). “Attention-Aware Age-Agnostic Visual Place Recognition.” 2019. Masters Thesis, Delft University of Technology. Accessed February 28, 2021. http://resolver.tudelft.nl/uuid:250d37a9-bc0d-4f8f-8d1a-d31a98dc22d7.

MLA Handbook (7th Edition):

Li, Jiahui (author). “Attention-Aware Age-Agnostic Visual Place Recognition.” 2019. Web. 28 Feb 2021.

Vancouver:

Li J(. Attention-Aware Age-Agnostic Visual Place Recognition. [Internet] [Masters thesis]. Delft University of Technology; 2019. [cited 2021 Feb 28]. Available from: http://resolver.tudelft.nl/uuid:250d37a9-bc0d-4f8f-8d1a-d31a98dc22d7.

Council of Science Editors:

Li J(. Attention-Aware Age-Agnostic Visual Place Recognition. [Masters Thesis]. Delft University of Technology; 2019. Available from: http://resolver.tudelft.nl/uuid:250d37a9-bc0d-4f8f-8d1a-d31a98dc22d7

.