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 27 for “"Conditional Generation"”.

  1. Conditional generation of temporally-ordered event sequences

    … these problems in a single model, a BART-based conditional generation model learned as a denoising autoencoder. This model operates over sequences of events represented as predicate-argument structures. At training, we take event sequences, shuffle them, delete events, and then attempt to …

    texas Repository record for Conditional generation of temporally-ordered event sequences (opens in a new tab)

  2. Conditional Generation of Protein Sequence and Structure

    … properly use these deep learning models for the generation of protein sequences and structures with user defined functional and biochemical properties. This was the focus of my dissertation work. I first interrogated this question by taking pre trained structure prediction networks, namely …

    washington Repository record for Conditional Generation of Protein Sequence and Structure (opens in a new tab)

  3. Enabling Conditional Generation Without Training: Plug & Play Generative Modelling using Diffusion Models

    … tasks, including text, image, video, and audio generation. By iteratively refining standard Gaussian noise into structured signals, these models offer remarkable flexibility. However, their practical capabilities remain constrained by the limited availability of large-scale annotated datasets, …

    jhu Repository record for Enabling Conditional Generation Without Training: Plug & Play Generative Modelling using Diffusion Models (opens in a new tab)

  4. Machine Learning for Data-Driven and Simulated Problems of Fluid Dynamics

    … limited predictability. The first addresses a conditional downscaling problem motivated by offshore wind energy, where daily large-scale atmospheric fields are used to predict hourly local variability along the central California coast. The second develops a parametric physics-informed …

    houston Repository record for Machine Learning for Data-Driven and Simulated Problems of Fluid Dynamics (opens in a new tab)

  5. Wavelet Flow: Fast Training of High Resolution Normalizing Flows

    … includes models of lower resolution signals and conditional generation of higher resolution signals, i.e., super resolution. A major advantage of Wavelet Flow is the ability to construct generative models for high resolution data (e.g., 1024 1024 images) that are impractical with previous models. …

    york Repository record for Wavelet Flow: Fast Training of High Resolution Normalizing Flows (opens in a new tab)

  6. Tree-based Generative Models with Applications to Microbiome Compositional Data

    <p>In this work, we study conditional generative models of microbiome compositional data. Specifically, we propose three approaches for the conditional generation of such data: (1) a parametric model called the "logistic-tree normal" (LTN) model, (2) a conditional normalizing flow based on …

    duke Repository record for Tree-based Generative Models with Applications to Microbiome Compositional Data (opens in a new tab)

  7. Impressions: Understanding Visual Semiotics and Aesthetic Impact

    … that existing multimodal image captioning and conditional generation models struggle to simulate plausible human responses to images. However, this dataset significantly improves their ability to model impressions and aesthetic evaluations of images through fine-tuning and few-shot adaptation.

    gatech Repository record for Impressions: Understanding Visual Semiotics and Aesthetic Impact (opens in a new tab)

  8. Latent Walking Techniques for Conditioning GAN-Generated Music

    Artificial music generation is a rapidly developing field focused on the complex task of creating neural networks that can produce realistic-sounding music. Generating music is very difficult; components like long and short term structure present time complexity, which can be difficult for neural …

    vt Repository record for Latent Walking Techniques for Conditioning GAN-Generated Music (opens in a new tab)

  9. Developing Novel in Silico Tools for Heterochiral Macrocycle Design

    … spanning conformational sampling, sequence generation, and target-specific design. First, a Multiple Structure Alignment (MStA) framework paired with an axial attention architecture is introduced for backbone conformation prediction, demonstrating that co-evolutionary information is not …

    oregon Repository record for Developing Novel in Silico Tools for Heterochiral Macrocycle Design (opens in a new tab) Embargoed until 2028-07-31

  10. Dangaus skliauto generavimas pasitelkiant neuroninius tinklus /

    … to generate high-resolution skydomes, a DDIM generation method was implemented to shorten the generation process, and a conditional generation method was implemented to control the generation of the model. The influence of variance schedules, number of generation steps and cascaded diffusion …

    vilnius Repository record for Dangaus skliauto generavimas pasitelkiant neuroninius tinklus / (opens in a new tab)

  11. Deep learning with simulated data for medical imaging

    … thesis explores innovative approaches to data generation and synthesis, aiming to augment existing datasets and create new ones. The work employs advanced techniques such as cross-modality synthesis using learned local attention masks to generate MRA images from T2-weighted brain MR images, …

    whiterose Download PDF for Deep learning with simulated data for medical imaging (opens in a new tab)

  12. Advancing Control: A Latent Variable Approach To Controllable Image Synthesis

    … models has predominantly focused on StyleGAN and conditional variational autoencoders (cVAEs). StyleGAN and similar style-based generators are considered relatively difficult to train due to instability associated with generative adversarial learning. cVAEs are generally more stable in training, …

    unsw Repository record for Advancing Control: A Latent Variable Approach To Controllable Image Synthesis (opens in a new tab)

  13. Full-Atom Peptide and Loop Structure Controllable Generations with Deep Generative Models

    … I developed methodologies for full-atom helix generation. HelixGAN, the first generative adversarial network for atomic-level helix design, enables conditional generation under structural constraints, providing a new tool for early-stage drug discovery. Building on this, HelixDiff, a …

    toronto-retro Repository record for Full-Atom Peptide and Loop Structure Controllable Generations with Deep Generative Models (opens in a new tab)

  14. Unsupervised Text Representation Learning with Interactive Language

    … We formulate the latent factor learning as a conditional generation process by dynamically querying a memory of latent mode vectors for template information that is shared across the data samples. A potential advantage of using the latent mode vectors is that they can make the resulting model …

    washington Repository record for Unsupervised Text Representation Learning with Interactive Language (opens in a new tab)

  15. From Completion to Generation: Learning Representation for 3D Content Enhancement and Creation

    … the capabilities of 3D content enhancement and generation. We begin by enhancing existing 3D content collected from the real world. 3D acquisition devices, such as LiDAR and RGB-D cameras, often capture 3D data as point clouds, which are typically sparse and incomplete. To improve the quality of …

    anu Repository record for From Completion to Generation: Learning Representation for 3D Content Enhancement and Creation (opens in a new tab)

  16. Learning factorised representation via generative models.

    … factorised latent space allows the model to do conditional generation. The results can achieve a state-of-the-art Fréchet inception distance (FID) score. Then we investigate the unsupervised object-centric representation learning, which can be seen as a deeper level of the foreground …

    soton Download PDF for Learning factorised representation via generative models. (opens in a new tab)

  17. Towards Effective and Controllable Neural Text Generation

    … achieved remarkable empirical success in text generation tasks. Users demand both effective and controllable generation, which mean respectively to generate high-quality human-level sequences of words, and to control text output in certain aspects to accommodate different practical needs. …

    buffalo Repository record for Towards Effective and Controllable Neural Text Generation (opens in a new tab)

  18. Guidance Conditions in Generative Modeling: Elevating Discriminative Capabilities and Controllability

    … various guidance methods have emerged for conditional generation. For example, text-to-image generation and image captioning utilize encoded features as conditions, while in-context learning conditions generation directly on the input data. Achieving this level of conditional adherence …

    buffalo Repository record for Guidance Conditions in Generative Modeling: Elevating Discriminative Capabilities and Controllability (opens in a new tab)

  19. Leveraging Generative Models for Music and Signal Processing

    … to synthesize or re-synthesize multimedia; conditional generative modeling empowers us to control the outputs of these models. By steering a generative model with conditioning information, we can sketch the essential aspects of our creative vision, and the generative model will fill in the …

    washington Repository record for Leveraging Generative Models for Music and Signal Processing (opens in a new tab)

  20. Docile Bodies or Resistant Bodies? Tracing Power Dynamics in Elite Iranian Muslim Sportswomen's Experiences

    … Islamic Republic values whose legitimacy remains conditional upon conformity to dominant gendered expectations. Surveillance operates through institutional monitoring, digital observation, interpersonal relations, and self-regulation, producing athletes who continuously negotiate the boundaries of …

    ottawa-retro Repository record for Docile Bodies or Resistant Bodies? Tracing Power Dynamics in Elite Iranian Muslim Sportswomen's Experiences (opens in a new tab)

Page 1 of 2