Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 108 for “"sequence modeling"”.

  1. Sequence Modeling with Linear Complexity

    Sequence modeling represents a fundamental task in machine learning. Today, non-autoregressive attention-based methods that use self-attention are deemed standard for modeling sequences. The major drawback of self-attention is the quadratic time complexity O(n^2) for the processing of a sequence …

    carleton Repository record for Sequence Modeling with Linear Complexity (opens in a new tab)

  2. Understanding stories via event sequence modeling

    Understanding stories, i.e. sequences of events, is a crucial yet challenging natural language understanding (NLU) problem, which requires dealing with multiple aspects of semantics, including actions, entities and emotions, as well as background knowledge. In this thesis, towards the goal of …

    uiuc Repository record for Understanding stories via event sequence modeling (opens in a new tab)

  3. Neural Probabilistic Methods for Event Sequence Modeling

    This thesis focuses on modeling event sequences, namely, sequences of discrete events in continuous time. We build a family of generative probabilistic models that is able to reason about what events will happen in the future and when, given the history of previous events. Under our models, each …

    jhu Repository record for Neural Probabilistic Methods for Event Sequence Modeling (opens in a new tab)

  4. Generalizing speech recognition techniques for monotonic sequence modeling applications

    <p>This thesis demonstrates how modeling techniques from speech recognition can be advantageous in a variety of other problems. The problems to which we apply these techniques include visual-only speech recognition (lipreading), restoration of characters missing from old Hawaiian orthography but …

    oxford Repository record for Generalizing speech recognition techniques for monotonic sequence modeling applications (opens in a new tab)

  5. Structured Sequence Modeling with Graphs, Flows, and Real-world Knowledge

    … gives rise to an important class of structures: sequences of events. In natural languages, we communicate with each other by linearizing our thoughts and turning them into a sequence of sound waves. In finance, investors make trade decisions based on the sequence of stock prices and of different …

    anu Repository record for Structured Sequence Modeling with Graphs, Flows, and Real-world Knowledge (opens in a new tab)

  6. SEQUENCE MODELING OF RAFT POLYMERIZATIONS WITH THE METHOD OF MOMENTS

    Attempts to model the sequence structure of copolymers consisted of probabilistic functions that were incomplete and inaccurate. A novel technique to track sequence parameters is developed that determines not only copolymer composition, but sequence distribution as well. RAFT polymerizations are …

    maryland Repository record for SEQUENCE MODELING OF RAFT POLYMERIZATIONS WITH THE METHOD OF MOMENTS (opens in a new tab)

  7. Neural Sequence Modeling for Domain-Specific Language Processing: A Systematic Approach

    In recent years, deep learning based sequence modeling (neural sequence modeling) techniques have made substantial progress in many tasks, including information retrieval, question answering, information extraction, machine translation, etc. Benefiting from the highly scalable attention-based …

    vt Repository record for Neural Sequence Modeling for Domain-Specific Language Processing: A Systematic Approach (opens in a new tab)

  8. ML and System Approaches for Data Processing Optimization and Long Sequence Modeling

    … challenges due to its quadratic dependency on sequence length. This limitation restricts their ability to handle and manage long data (e.g., large tables, videos, and audio). To enable LLMs ability to process long sequence inputs, we investigate whether linear Recurrent Neural Networks (RNNs) …

    cornell Repository record for ML and System Approaches for Data Processing Optimization and Long Sequence Modeling (opens in a new tab)

  9. A study of Bluetooth Frequency Hopping sequence : modeling and a practical attack

    … this work, we investigate the Frequency Hopping sequence and specifically the hop selection kernel. We analyze the operation of the kernel hardware by partitioning it into three parts. Based on this modeling, we propose an attacking method for the hop selection kernel. The proposed method shows …

    msu Repository record for A study of Bluetooth Frequency Hopping sequence : modeling and a practical attack (opens in a new tab)

  10. IMPLICATIONS OF STRUCTURAL BIAS AND COMPATIBILITY: AN APPROXIMATION ANALYSIS OF SEQUENCE MODELING ARCHITECTURES

    … the theoretical aspects of the approximation of sequence modeling architectures, elucidating a range of prominent neural network architectures, RNNs, RNN encoder-decoders, TCNs and Transformers. The essence of this study lies in the underlying assumption that conducting a rigorous analytical …

    nus Repository record for IMPLICATIONS OF STRUCTURAL BIAS AND COMPATIBILITY: AN APPROXIMATION ANALYSIS OF SEQUENCE MODELING ARCHITECTURES (opens in a new tab) Download file for IMPLICATIONS OF STRUCTURAL BIAS AND COMPATIBILITY: AN APPROXIMATION ANALYSIS OF SEQUENCE MODELING ARCHITECTURES (opens in a new tab)

  11. Enhancing representation learning with tensor decompositions for knowledge graphs and high dimensional sequence modeling

    … knowledge graph data and high dimensional sequences. In the first part of this thesis, we first review multiple relational learning models based on tensor decomposition for knowledge graphs. We point out that relational learning is in fact a means of learning representations through one-hot …

    lmu-germany Repository record for Enhancing representation learning with tensor decompositions for knowledge graphs and high dimensional sequence modeling (opens in a new tab)

  12. Engineering Antimicrobial Proteins and Peptides as Next-Generation Antimicrobials via Genome-mining, High-throughput Functional Screening, and Sequence Modeling

    … for informing protein design. This work spans sequence-function mapping of two antimicrobial protein families and the development of efficient strategies for their continued engineering. First, genome-mining approaches are applied to investigate the small antimicrobial protein family of class …

    umn Repository record for Engineering Antimicrobial Proteins and Peptides as Next-Generation Antimicrobials via Genome-mining, High-throughput Functional Screening, and Sequence Modeling (opens in a new tab)

  13. State Space Model Based Channel Estimation

    … can be viewed as a time series and hence various sequence modeling approaches can be used to perform channel estimation. Traditional sequence modeling approaches like recurrent neural network and transformers have a disadvantage of high computational complexity. In the recent years, many state …

    tamu Repository record for State Space Model Based Channel Estimation (opens in a new tab)

  14. On the Neural Architecture of Language Models

    Modern large-scale sequence modeling requires architectures that can represent long-range dependencies while remaining computationally efficient. Transformers dominate many sequence modeling tasks because self-attention enables expressive and highly parallel computation, but its quadratic time and …

    ottawa-retro Repository record for On the Neural Architecture of Language Models (opens in a new tab)

  15. Neural Network vs. Rule-Based G2P: A Hybrid Approach to Stress Prediction and Related Vowel Reduction in Bulgarian

    … in G2P tasks. We investigate neural network sequence-to-sequence modeling for the prediction of syllable stress and resulting vowel reductions in the Bulgarian language. We then develop a hybrid G2P approach which combines manually written grapheme-to-phoneme mapping rules with neural …

    cuny-grad Repository record for Neural Network vs. Rule-Based G2P: A Hybrid Approach to Stress Prediction and Related Vowel Reduction in Bulgarian (opens in a new tab)

  16. Novel Methods for Natural Language Modeling and Pretraining

    This thesis is about modeling language sequences to achieve lower perplexity, better generation, and benefit downstream language tasks; specifically, this thesis addresses the importance of natural language features including the segmentation feature, lexical feature, and alignment feature. We …

    waterloo Repository record for Novel Methods for Natural Language Modeling and Pretraining (opens in a new tab)

  17. Performance profiling and activity-based state pruning for efficient Mamba inference

    … to Transformer-based architectures for sequence modeling, thanks to their favorable scaling behavior and efficiency on long-context tasks. Among these models, Mamba has attracted significant attention for its linear-time sequence modeling capability. However, the internal performance …

    iastate Repository record for Performance profiling and activity-based state pruning for efficient Mamba inference (opens in a new tab)

  18. Reverse Engineering an Expository Paragraph for Students with Mild to Moderate Disabilities and English Learners

    … how experts approach a task, typically in a sequence of beginning, middle, and ending steps. This intervention applied reverse engineering theory to the modelling step of paragraph writing instruction. The link between the prewriting and drafting steps of the expository paragraph writing …

    unlv Repository record for Reverse Engineering an Expository Paragraph for Students with Mild to Moderate Disabilities and English Learners (opens in a new tab) Download PDF for Reverse Engineering an Expository Paragraph for Students with Mild to Moderate Disabilities and English Learners (opens in a new tab)

Page 1 of 6