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 55 for “"Flow matching"”.
-
Associative Memory for Hyperdimensional Spatial Representations with Geodesic Flow Matching
… entirely within the embedding space. Geodesic Flow Matching is introduced as a model that learns a continuous time-dependent velocity field to transport corrupted representations back to the valid data manifold. Standard Euclidean Flow matching is shown to be insufficient for high-dimensional …
-
Video Generation Techniques for Novel View Synthesis with Flow-Matching Transformers
… conditioning. In this thesis, we introduce MVFlow, a new NVS model that extends this approach to a different image generation architecture: a flow-matching diffusion transformer, specifically FLUX.1, which has demonstrated strong performance in image synthesis. We evaluate MVFlow under varying …
-
Capacity dynamics of feed-forward, flow-matching networks exposed to random disruptions
… of capacity disruptions in complex multi-tier flow-matching feed-forward networks (FFN). In particular, we relax basic structural assumptions of FFN, introduce random propagation times, study the impact of inventory buffers on propagation times, and make initial efforts to model random network …
-
Conditional flow matching for fast generation of LES-quality instantaneous urban microclimate fields
… a generative surrogate model using Conditional Flow Matching (CFM) with optimal transport to reconstruct and generate plausible instantaneous multi-variable urban microclimate snapshots. The model learns a time-dependent vector field to transport Gaussian noise into physically realistic flow …
-
Information-Geometric Generative Models for Mixed Continuous–Categorical Tabular Data
… problem for continuous generative models. Flow Matching and Gaussian diffusion operate in Euclidean spaces, whereas categorical laws live on probability simplices and can be highly imbalanced. We propose a logit-coordinate framework that embeds categorical variables in smoothed …
-
Energy-based models for speech synthesis
… them to related frameworks such as diffusion and flow-matching methods. This thesis presents a study on applying energy-based models for speech synthesis. The primary objective of this research is to train energy-based TTS models with commonly used noise contrastive estimation (NCE). During this …
-
Not All Policies Are Equal: Training Flow-Based Policies for Efficient Downstream Steering
Generative policies based on diffusion and flow matching can represent multimodal robotaction distributions and have become a practical foundation for visuomotor imitation learning. However, these policies are usually trained as action generators rather than as controllable latent- action maps. …
-
Deep generative modeling for protein structure prediction and design
… design. To address this need, I developed FlowPacker, which applies torsional flow matching and equivariant graph neural networks for protein side-chain packing. I show that FlowPacker attains state-of-the-art performance in both runtime and accuracy. Moreover, I show that unlike other …
-
Conformal Prediction for Time-Series and Flow-Based Generative Models
… 1 and 2) and modeling data distributions with flow-based generative models (Chapters 3 and 4). Chapter 1 extends conformal prediction to time-series data, providing prediction intervals with bounded conditional coverage gaps and demonstrating superior empirical performance. Chapter 2 builds on …
-
Trajectory Modeling using Generative Approaches for Scheduling, Planning, and Multi-Agent Systems
… adversarial tracking models, (2) creation of a flow-matching framework for trajectory prediction and generation, (3) a multi-agent reinforcement learning algorithm for large, sparse partially observable domains, (4) a kinematically constrained gradient guidance method for learning diverse, agile …
-
Education and the labour market
… of random search, directed search, and stock-flow matching differ substantially in the way they assume that job seekers and firms behave during the recruitment process in these respects and we conjecture that our findings are most readily explained by stock-flow matching models - if these were …
-
Diffusion Models for Protein Structure Design: From Backbone Generation to Atomic-Resolution Enzyme Design
… molecules. Finally, we present RFdiffusion2, a flow-matching model trained from random weight initializations capable of unindexed atomic motif scaffolding, enabling the design of enzymes with complex active sites. In all cases we validate the design capabilities of the models \textit{in vitro}. …
-
Advances in Bayesian Computation: Bridging Modern Machine Learning and Traditional Monte Carlo Methods
… Bayesian computation method that integrates flow matching with Markovian sampling, enhancing the exploration of complex target distributions through adaptive tempering mechanisms. Our work extends Bayesian nonparametric approaches to linear regression models, effectively handling outliers and …
Page 1 of 3