- Faculty of Arts and Sciences
- FAS Theses and Dissertations
- Communities & Collections
- By Issue Date
- FAS Department
- Quick submit
- Waiver Generator
- DASH Stories
- Accessibility
- COVID-related Research
Terms of Use
- Privacy Policy
- By Collections
- By Departments
Building the Theoretical Foundations of Deep Learning: An Empirical Approach
Citable link to this page
Collections.
- FAS Theses and Dissertations [6138]
Contact administrator regarding this item (to report mistakes or request changes)
- Help & FAQ
PhD thesis: Foundations and advances in deep learning
- Computer Science
Research output : Book/Report › Other report
T1 - PhD thesis
T2 - Foundations and advances in deep learning
AU - Cho, Kyunghyun
M3 - Other report
BT - PhD thesis
PB - Aalto University
Help | Advanced Search
Computer Science > Computer Vision and Pattern Recognition
Title: master's thesis : deep learning for visual recognition.
Abstract: The goal of our research is to develop methods advancing automatic visual recognition. In order to predict the unique or multiple labels associated to an image, we study different kind of Deep Neural Networks architectures and methods for supervised features learning. We first draw up a state-of-the-art review of the Convolutional Neural Networks aiming to understand the history behind this family of statistical models, the limit of modern architectures and the novel techniques currently used to train deep CNNs. The originality of our work lies in our approach focusing on tasks with a low amount of data. We introduce different models and techniques to achieve the best accuracy on several kind of datasets, such as a medium dataset of food recipes (100k images) for building a web API, or a small dataset of satellite images (6,000) for the DSG online challenge that we've won. We also draw up the state-of-the-art in Weakly Supervised Learning, introducing different kind of CNNs able to localize regions of interest. Our last contribution is a framework, build on top of Torch7, for training and testing deep models on any visual recognition tasks and on datasets of any scale.
Submission history
Access paper:.
- Other Formats
References & Citations
- Google Scholar
- Semantic Scholar
DBLP - CS Bibliography
Bibtex formatted citation.
Bibliographic and Citation Tools
Code, data and media associated with this article, recommenders and search tools.
- Institution
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .
Navigation Menu
Search code, repositories, users, issues, pull requests..., provide feedback.
We read every piece of feedback, and take your input very seriously.
Saved searches
Use saved searches to filter your results more quickly.
To see all available qualifiers, see our documentation .
- Notifications
Rex Ying's Ph.D. Thesis, Stanford University
RexYing/Rex-Ying-Thesis
Folders and files, repository files navigation, rex ying's thesis.
Here you could find Rex Ying's Ph.D. Thesis at Stanford University, advised by Jure Leskovec.
Towards Expressive and Scalable Deep Representation Learning for Graphs
My Ph.D. Thesis revolves around graph representation learning. Specifically, we discuss the use of expressive, scalable and explainable graph neural networks on structured data, and demonstrate its applications in social platforms , biological networks and physical simulations.
Got any suggestions?
We want to hear from you! Send us a message and help improve Slidesgo
Top searches
Trending searches
holy spirit
36 templates
39 templates
memorial day
12 templates
21 templates
ai technology
169 templates
11 templates
Thesis Defense Presentation templates
If you are ready to defend your dissertation, our free thesis defense ppt templates and google slides themes are a great tool for you to impress the committee and get that desired degree you have always wanted..
It seems that you like this template!
Greenwashing impact thesis defense.
Download the Greenwashing Impact Thesis Defense presentation for PowerPoint or Google Slides. Congratulations, you have finally finished your research and made it to the end of your thesis! But now comes the big moment: the thesis defense. You want to make sure you showcase your research in the best way...
Premium template
Unlock this template and gain unlimited access
Chemistry Thesis
Everything in this world is composed of atoms and, within science, chemistry is what studies these. Thanks to chemists we can understand what our universe is made of. As a graduate student who has chosen this career path, you’re about to finish your master’s program, but only the thesis defense...
US Colonial History: The 13 Colonies Thesis Defense
Download the US Colonial History: The 13 Colonies Thesis Defense presentation for PowerPoint or Google Slides. Congratulations, you have finally finished your research and made it to the end of your thesis! But now comes the big moment: the thesis defense. You want to make sure you showcase your research...
Psychology of Aroma and Scent Thesis Defense
Download the Psychology of Aroma and Scent Thesis Defense presentation for PowerPoint or Google Slides. Congratulations, you have finally finished your research and made it to the end of your thesis! But now comes the big moment: the thesis defense. You want to make sure you showcase your research in...
Raven Mythology and Folklore Thesis Defense
Download the Raven Mythology and Folklore Thesis Defense presentation for PowerPoint or Google Slides. Congratulations, you have finally finished your research and made it to the end of your thesis! But now comes the big moment: the thesis defense. You want to make sure you showcase your research in the...
East Asian Studies Thesis
There’s an academic discipline that covers the history, language, culture and other aspects of countries such as China, Japan or Korea. Are you studying a master’s in East Asian Studies and need help with the presentation for the thesis defense? Try our education template!
Child Care Thesis
Taking care of kids is not an easy task at times, but it’s quite rewarding! In fact, doing research on how to take care of them will be key for a better future. Present your thesis on child care with this lovely free template!
Simple Style Thesis Defense
We know how much work involves completing a thesis, and that is why it is also important that you get to impress the committee when making your final presentation. If you are looking for something simple but at the same time formal, take a look at this proposal from Slidesgo....
Humanization in Healthcare Thesis Defense
Download the Humanization in Healthcare Thesis Defense presentation for PowerPoint or Google Slides. Congratulations, you have finally finished your research and made it to the end of your thesis! But now comes the big moment: the thesis defense. You want to make sure you showcase your research in the best...
Elegant Bachelor Thesis
Present your Bachelor Thesis in style with this elegant presentation template. It's simple, minimalist design makes it perfect for any kind of academic presentation. With an array of features such as section dividers, images, infographics and more, you can easily create a professional and creative presentation that stands out from...
Elegant Black & White Thesis Defense
Present your research findings with grace and assertiveness through this template. Available for Google Slides and PowerPoint, this design set offers minimalistic charm with its simple, gray scale elegance. The template not only provides a polished platform to showcase your thesis but also ensures seamless and efficient delivery of your...
Monarch Butterfly Thesis Defense Infographics
Download the Monarch Butterfly Thesis Defense Infographics template for PowerPoint or Google Slides and discover the power of infographics. An infographic resource gives you the ability to showcase your content in a more visual way, which will make it easier for your audience to understand your topic. Slidesgo infographics like...
Effects Of Decolonization Thesis Defense
Download the Effects Of Decolonization Thesis Defense presentation for PowerPoint or Google Slides. Congratulations, you have finally finished your research and made it to the end of your thesis! But now comes the big moment: the thesis defense. You want to make sure you showcase your research in the best...
Egyptian History Thesis
Egyptian history is always a fascinating topic, full of mysteries and secrets. If your thesis is about this and you’re looking for a good way to present your findings clearly, why don’t you give this great template by Slidesgo a try?
Dissertation Defense - Doctor of Philosophy (Ph.D.) in History
Download the "Dissertation Defense - Doctor of Philosophy (Ph.D.) in History" presentation for PowerPoint or Google Slides. As university curricula increasingly incorporate digital tools and platforms, this template has been designed to integrate with presentation software, online learning management systems, or referencing software, enhancing the overall efficiency and effectiveness of...
Cellular Respiration and its Impact on Health Research Thesis Defense
Download the "Cellular Respiration and its Impact on Health Research Thesis Defense" presentation for PowerPoint or Google Slides. Congratulations, you have finally finished your research and made it to the end of your thesis! But now comes the big moment: the thesis defense. You want to make sure you showcase...
After months or even years of researching about art and a lot of hard work, you’ve at last finished your master’s thesis. Now it’s time for the crucial part: the defense. Impress the committee and get that degree with the help of Slidesgo!
Minimalist Thesis Defense
Get rid of all that nervousness and enter the room full of confidence and willing to impress everyone in your thesis defense. Wait, first you must customize this new template so that your presentation is ready! Nothing feels as boring to the assessment committee than overcrowded slides, so our proposal...
- Page 1 of 43
Great presentations, faster
Slidesgo for Google Slides :
The easy way to wow
Register for free and start editing online
COMMENTS
The objective of this thesis was to study the application of deep learning in image classification using convolutional neural networks. The Python programming language with the TensorFlow framework and Google Colaboratory hardware were used for the thesis. Models were chosen from available ones online and adjusted by the author.
In July this year I finally defended my PhD which mainly focused on (adversarial) robustness and uncertainty estimation in deep learning. In my case, the defense consisted of a (public) 30 minute talk about my work, followed by questions from the thesis committee and audience. In this article, I want to share the slides and some lessons learned in preparing for my defense.
Go 1.0 was released in March 2012 [22]. The focus of this thesis is to integrate GPU computation with the Go language for the purpose of developing deep learning models. This chapter includes a review of some of the packages that were developed for GPU computation with Go, the applications that use them, and other deep learning frameworks. 2.1 ...
challenging while blur is nonuniform. Since the rise of deep learning, many recent - approaches are based on Convolutional Neural Networks (CNNs).These CNN-based approaches are diverse, in terms of their structures and components. However, existing methods have many disadvantages, for instance, they require intensive computation
The new model family introduced in this thesis is summarized under the term Recursive Deep Learning. The models in this family are variations and extensions of unsupervised and supervised recursive neural networks (RNNs) which generalize deep and feature learning ideas to hierarchical structures. The RNN models of this thesis
Abstract Deeplearninghasattractedtremendousattentionfromresearchersinvariousfieldsof informationengineeringsuchasAI,computervision,andlanguageprocessing[Kalch-
Dismiss. 1 Introduction to Deep Learning Mustafa Mustafa NERSC @mustafa240m Data Seminars, NERSC March 2019, Berkeley Lab 2 @mustafa240m. 2. Deep Learning is powering many recent technologies 3 @mustafa240m Deep Learning is powered by Deep Neural Networks. 3.
Her ambition and foresight ignited my passion for bridging the research in deep learning and hardware. Sitting on the same floor with Fei-Fei and her students spawned many researchspark. IsincerelythankFei-Fei'sstudentsAndrejKarpathy,YukeZhu,JustinJohnson,
During the PhD course, I explore and establish theoretical foundations for deep learning. In this thesis, I present my contributions positioned upon existing literature: (1) analysing the generalizability of the neural networks with residual connections via complexity and capacity-based hypothesis complexity measures; (2) modeling stochastic ...
largest photo sharing services, that use Deep Learning technologies to e ciently order and sort out piles of pictures 2, to better target advertising, or to nd people associated to faces [45]. Startups also are using Deep Learning to build better recognition products and to revolutionize the market providing new services 3. Furthermore, it is ...
In this thesis, we take a ``natural sciences'' approach towards building a theory for deep learning. We begin by identifying various empirical properties that emerge in practical deep networks across a variety of different settings. Then, we discuss how these empirical findings can be used to inform theory. Specifically, we show the following ...
An overview of Convolutional Neural Network (CNN) and its applications in deep learning.
This is a deep learning presentation based on Deep Neural Network. It reviews the deep learning concept, related works and specific application areas.It describes a use case scenario of deep learning and highlights the current trends and research issues of deep learning. Read more. Technology. 1 of 79.
Deep learning (DL), a branch of machine learning (ML) and artificial intelligence (AI) is nowadays considered as a core technology of today's Fourth Industrial Revolution (4IR or Industry 4.0). Due to its learning capabilities from data, DL technology originated from artificial neural network (ANN), has become a hot topic in the context of computing, and is widely applied in various ...
Deep learning models have been used for this task, and traditional methods such as clustering and rule-based systems are widely used as well. This thesis aims to compare deep learning models with traditional algorithms for anomaly detection in network traffic and analyze the trade-offs between the models in terms of accuracy and scalability. 19.
greedy layerwise learning (Belilovsky et al. 2019). Here we established a new mechanism for biologically plausible neural networks by combining the feedback alignment update and the layerwise learning. Furthermore, with the randomized layerwise learning we proposed, we show that convolutional neural networks can be trained without end-to-
In this chapter, we will mak e an overview of what exists in terms of deep learning toolset. W e will define four deep learning compilers (Tiramisu, TVM, Glow and XLA) and five differen t ...
TY - BOOK. T1 - PhD thesis. T2 - Foundations and advances in deep learning. AU - Cho, Kyunghyun. PY - 2014. Y1 - 2014. M3 - Other report. BT - PhD thesis
PhD Dissertations [All are .pdf files] Probabilistic Reinforcement Learning: Using Data to Define Desired Outcomes, and Inferring How to Get There Benjamin Eysenbach, 2023. Data-driven Decisions - An Anomaly Detection Perspective Shubhranshu Shekhar, 2023. METHODS AND APPLICATIONS OF EXPLAINABLE MACHINE LEARNING Joon Sik Kim, 2023. Applied Mathematics of the Future Kin G. Olivares, 2023
The goal of our research is to develop methods advancing automatic visual recognition. In order to predict the unique or multiple labels associated to an image, we study different kind of Deep Neural Networks architectures and methods for supervised features learning. We first draw up a state-of-the-art review of the Convolutional Neural Networks aiming to understand the history behind this ...
This thesis considers deep learning theories of brain function, and in particular biologically plausible deep learning. The idea is to treat a standard deep network as a high-level model of a neural circuit (e.g., the visual stream), adding biological constraints to some clearly artificial features. Two big questions are possible. First,
Towards Expressive and Scalable Deep Representation Learning for Graphs. My Ph.D. Thesis revolves around graph representation learning. Specifically, we discuss the use of expressive, scalable and explainable graph neural networks on structured data, and demonstrate its applications in social platforms , biological networks and physical ...
Download the "Dissertation Defense - Doctor of Philosophy (Ph.D.) in History" presentation for PowerPoint or Google Slides. As university curricula increasingly incorporate digital tools and platforms, this template has been designed to integrate with presentation software, online learning management systems, or referencing software, enhancing ...