Artificial Intelligence in Progress

Home   Artificial Intell.   Bio-exact   Chemistry   Computer Sci. BSc   Earth Sciences   Mathematics   Physics & Astr.   Science Educ.   Submit   Webmaster        
bachelors   masters   all  


Displaying theses 1-10 of 504 total
Previous  1  2  3  4  5  6  7  8  9  10  11  Next  Search:

K.W. Korrel
Master programme: Artificial Intelligence November 26th, 2018
Institute: ILLC Research group: Language and Computation Graduation thesis Supervisor: Dieuwke Hupkes
From Sequence to Attention; Search for a Compositional Bias in Sequence-to-Sequence Models
Although sequence-to-sequence models have successfully been applied to many tasks, they are shown to have poor compositional skills. The principle of compositionality states that the meaning of a complex expression is a function only of its constituents and the manner in which they are combined. When a model would thus have an understanding of the individual constituents and can combine them in novels ways, this would allow for efficient learning and generalization. We first develop Attentive Guidance to show that guiding a sequence-to-sequence model in its attention modeling can help it find disentangled representations of the input symbols and to process them individually. Later we develop the sequence-to-attention architecture, a new model for sequence-to-sequence tasks with more emphasis on sparse attention modeling. We show that this architecture can find similar compositional solutions as can be developed with Attentive Guidance, without requiring attention annotations in the training data.
picture that illustrates the research done
Scientific abstract (pdf 2K)   Full text (pdf 1450K)

L.J. Pascha
Master programme: Artificial Intelligence November 26th, 2018
Institute: Informatics Institute Research group: QUVA Lab Graduation thesis Supervisor: Efstratios Gavves
Improving Word Embeddings for Zero-Shot Event Localisation by Combining Relational Knowledge with Distributional Semantics
Temporal event localisation of natural language text queries is a novel task in computer vision. Thus far, no consensus has been reached on how to predict the temporal boundaries of action segments precisely. While most attention in literature has been dedicated towards the representation of vision, here we attempt to improve the representation of language for event localisation by applying Graph Convolutions (GraphSAGE) on ConceptNet with distributional node embedding features. We argue that due to the large vocabulary size of language and currently small temporally sentence annotated datasets in scale and size, a high dependency is placed upon zero-shot performance. We hypothesise that our approach leads to more visually centred and structured language embeddings beneficial for this task. To test this, we design a wide-scale zero-shot dataset based on ImageNet to optimise our embeddings on and compare to other language embedding methods. State-of-the-art results are obtained on 5/17 popular intrinsic evaluation benchmarks, but with slightly lower performance on the TACoS dataset. Due to the almost complete overlap in train- and testset vocabulary, we deem additional testing necessary on a dataset that places more emphasis on word-relatedness; hypernyms, hyponyms and synonyms, which arguably makes language representation learning difficult.
picture that illustrates the research done
Scientific abstract (pdf 1K)   Full text (pdf 8953K)

T.E. Koenen
Master programme: Artificial Intelligence November 22nd, 2018
Institute: VU / Other Research group: Knowledge Representation and Reasoning Graduation thesis Supervisor: Peter Bloem
Text Generation and Annotation with Joint Multimodal Variational Autoencoders
Joint Multimodal Variational autoencoders are here applied to annotated text data and used to create a generative model over the two separate domains as well as allowing for cross domain mapping, by encoding mono-modal data and decoding the latent variable with the joint decoder.
picture that illustrates the research done
Scientific abstract (pdf 1K)   Full text (pdf 1784K)

J. Köhler
Master programme: Artificial Intelligence November 21st, 2018
Institute: Other Research group: Max Planck Institute for Intelligent Systems / EI Graduation thesis Supervisor: Efstratios Gavves
photo of the author
Differentially private data release by optimal compression
In this thesis we study how the principle of compressing data sets in an optimal way according to a measure of utility yields a mechanism that can be used to hide individuals within the set while still maintaining statistical usefulness for analysis tasks. As the problem of optimal compression is inherently difficult, we study two tractable instances that allow analytic sampling and analysis. Both approaches are further evaluated for their usefulness in practical applications. Finally, we sketch, how this work could be extended to less constrained assumptions on data set or utility measures.
picture that illustrates the research done
Scientific abstract (pdf 1K)   Full text (pdf 794K)

S.G.J. Bouwmeester
Master programme: Artificial Intelligence October 12th, 2018
Institute: ILLC Research group: Dialogue Modelling Group Graduation thesis Supervisor: Raquel Fernandez
photo of the author
Analysing Seq-to-seq Models in Goal-oriented Dialogue: Generalising to Disfluencies.
Data-driven dialogue systems are still far from understanding natural dialogue. Several aspects of natural language make it hard to capture in a system, such as unpredictability, mistakes and the width of the domain. In this thesis we take a step towards more natural data by examining disfluencies (i.e. mistakes). We test sequence to sequence models with attention on goal-oriented dialogue. Sequence to sequence models were chosen to overcome the unknown aspect of the mistakes, since they are known for their ability to generalise to unseen examples. The models are tested on disfluent dialogue data, the bAbI+ task, in addition to normal goal-oriented dialogue data, the bAbI task. In contrast to previous findings with memory networks, we find that the sequence to sequence model performs both the bAbI tasks as the bAbI+ task well achieving near perfect scores on both tasks. A slight decrease in performance is noticed when introducing disfluencies only to test data, only 80% accuracy is measured in this condition. This is surprising because memory networks are very similar to sequence to sequence models with attention.
picture that illustrates the research done
Scientific abstract (pdf 2K)   Full text (pdf 2316K)

F. Ambrogi
Master programme: Artificial Intelligence September 28th, 2018
Institute: Informatics Institute Research group: Computer Vision Graduation thesis Supervisor: Arnoud Visser
Evolving a Spiking Neural Network controller for low gravity environments
Development of an efficient and quickly trainable SNN controller for a legged rover, to operate in low gravity conditions. Several Evolutionary Algorithms were used to optimize the controller. Simulations are performed on MuJoCo, with the OpenAI Gym interface. The architecture is tested on some of its benchmarks, and then run on the low-g environment purposely created.
picture that illustrates the research done
Scientific abstract (pdf 2K)   Full text (zip 14843K)

L. Simonetto
Master programme: Artificial Intelligence September 28th, 2018
Institute: UvA / Other Research group: UvA / Other Graduation thesis Supervisor: E. Gavves
photo of the author
Generating spiking time series with Generative Adversarial Networks: an application on banking transactions
The task of data generation using Generative Models has recently gained more and more attention from the scientific community, as the number of applications in which these models work surprisingly well is constantly increasing. Some examples are image and video generation, speech synthesis and style transfer, pose guided image generation, cross-domain transfer and super resolution. Contrarily to such tasks generating data coming from the banking domain poses a different challenge, due to its atypical structure when compared with traditional data and its limited availability due to privacy restrictions. In this work, we analyze the feasibility of generating spiking time series patterns appearing in the banking environment using Generative Adversarial Networks. We develop a novel end-to-end framework for training, testing and comparing different generative models using both quantitative and qualitative metrics. Finally, we propose a novel approach that combines Variational Autoencoders with Generative Adversarial Networks in order to learn a loss function for datasets in which good similarity metrics are difficult to define.
picture that illustrates the research done
Scientific abstract (pdf 1K)   Full text (pdf 3178K)

D. Solis Morales
Master programme: Artificial Intelligence September 28th, 2018
Institute: Informatics Institute Research group: Information and Language Processing Systems Graduation thesis Supervisor: Ilya Markov
photo of the author
Transformer model for query suggestion
Query suggestions are query proposals after one or more queries have been submitted. They help users refine their queries when using a search engine. So far, query suggestion models have focused on complex recurrent encoder-decoder architectures to solve the task. Such complex architectures require a great amount of computational power and hours to train. This thesis proposes a novel neural model that reduces the complexity of current state-of-the-art models by using an encoder-decoder architecture that is based solely on attention. For this, it uses a Transformer model, that was first introduced for neural machine translation task, and it was shown to outperform state-of-the-art techniques. The AOL dataset is used to compare the model proposed with current state-of-the-art query suggestion models. Experiments show that it is possible to use a Transformer architecture for query suggestion task. This opens the door for future work to explore many different variants of Transformer models that are novel in the field of query suggestion task.
picture that illustrates the research done
Scientific abstract (pdf 73K)   Full text (pdf 968K)

N. Savov
Master programme: Artificial Intelligence September 26th, 2018
Institute: Informatics Institute Research group: Computer Vision Graduation thesis Supervisor: Sezer Karaoglu
photo of the author
Aiding Age Estimation with 3D Facial Geometry
In this work, we present deep network architectures that aim to improve age prediction by making use of 3D facial geometry. The exploited 3D geometry features were learned by a model which predicts a 3D face model from a 2D image. In addition, we propose a novel class distance loss for age estimation that penalizes high probability of distanced classes from the ground truth one. This allows age estimation to learn age-related features for a small interval around the ground truth age. We show that monocular 3D face reconstruction implicitly learns 3D facial geometry features that are age representative. Jointly learning age and 3D facial geometry lead to a significant improvement in the accuracy of age estimation. The sources of improvement are identified to be the features from monocular face reconstruction and the capability of multi-task learning. The proposed model helps to give better predictions the most for extreme poses and intensive expression.
picture that illustrates the research done
Scientific abstract (pdf 1K)   For more info or full text, mail to: S.Karaoglu@uva.nl

T.J. van Rozendaal
Master programme: Artificial Intelligence September 21st, 2018
Institute: UvA / Other Research group: UvA / Other Graduation thesis Supervisor: Taco Cohen
photo of the author
ASR Personalization by Voice Conversion
Automated Speech Recognition (ASR) systems are increasingly embedded into our personal lives. We use speech recognition in a variety of personal devices, such as smartphones or smart speakers. On these personal devices, the ASR system is mainly used by a single speaker. The problem of making an ASR system work better for a specific speaker is called ASR personalization. In this thesis, we investigate if voice conversion can be used to personalize an ASR system. Voice conversion is the task of converting a speech sample so that it sounds like a different speaker is saying the same sentence. We use voice conversion to change the speaker of an ASR training data set, and use the converted data to train a personalized ASR model. We show that voice conversion can be used to increase general ASR performance, but fail to personalize for one specific speaker in our setup.
picture that illustrates the research done
Scientific abstract (pdf 3K)   Full text (pdf 4478K)

Previous  1  2  3  4  5  6  7  8  9  10  11  Next  

This page is maintained by thesis@science.uva.nl