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 25 for “"Hypernetworks"”.
-
Hypernetworks Analysis of RoboCup Interactions
… to go beyond networks to multidimensional hypernetworks. In this context, this thesis investigates (i) is it possible to identify patterns of play, that lead a team to obtain an advantage ?, (ii) is it possible to forecast with a good degree of accuracy if a certain game action or sequence …
-
Capability-Aware Shared Hypernetworks for Heterogeneous Multi-Agent Coordination
… and meta-RL, we propose Capability-Aware Shared Hypernetworks (CASH), a new soft weight sharing architecture for heterogeneous coordination that use hypernetworks to explicitly reason about continuous agent capabilities in addition to local observations. Intuitively, CASH allows the team to learn …
-
Representations in zero-shot meta-reinforcement learning
… this, in Chapter 4, we propose the use of hypernetworks, i.e., neural networks that generate the weights and biases for other neural networks. In order to train hypernetworks stably, we introduce a novel initialization scheme that enables hypernetworks to improve generalization performance. …
-
Scalable Optimization of Binary Neural Networks Using Variational Quantum Algorithms
… of VQAs to train simulated quantum circuits as hypernetworks that generate effective BiNN configurations. By developing approximate methods for calculating the VQA objective and gradient, I significantly scale quantum hypernetworks—over two orders of magnitude larger than previous capabilities, …
-
Artificial Neural Networks with Dynamic Connections
… issue using the recently introduced concept of "hypernetworks", neural networks that generate other neural networks. The first part of this thesis discusses how these models can be used to learn dynamic features from unlabeled image and video data. The second part introduces Active Predictive …
-
Automated deep learning architecture design using differentiable architecture search (DARTS)
… algorithms rely on additional controllers and hypernetworks to generate architecture encoding or predict weights for sampled architectures. These controllers and hypernetworks might require optimal structure when deployed on a new task on a new dataset. And since this is done by hand, the …
-
Automatic Multilevel Feature Abstraction in Adaptable Machine Vision Systems
… The research investigated the application of Hypernetworks theory to building a multilevel, generally-applicable representation, through repeated application of a fundamental 'self-similarity' principle, that parts of objects assembled under a particular relation at one level, form whole …
-
Intelligent Knowledge Transfer for Multi-Stage and Multi-Task Learning
… a novel method named HyperPrompt that utilizes HyperNetworks to generate task-conditioned prompts for multi-task learning, where the task-specific knowledge can be flexibly shared via the HyperNetworks. • To alleviate the negative transfer problem from the perspective of gradient magnitudes, we …
-
Adaptable Policies and Accelerated Infrastructure for Learning for Heterogeneous Multi-Robot Coordination
… middle ground approach: Capability Aware Shared Hypernetworks (CASH). CASH is a soft weight sharing architecture that uses hypernetworks to efficiently learn a flexible shared policy that dynamically adapts to each robot post training. CASH outperforms baseline architectures in terms of …
-
Rethinking Maximum Likelihood Estimation
… and implemented in deep learning settings with hypernetworks. Inspired by this method, I propose a new grading scheme that incentives students to report their true beliefs, and show how the proposed method evaluates cognition and metacognition simultaneously. This thesis justifies both methods …
-
Achieving Flexible Fairness and Privacy in Federated Learning
… we performed the following: 1) We developed Fair HyperNetworks (FHN) -- a personalized federated learning algorithm based on hypernetworks that allows each client in the federation to choose which type of demographic fairness to enforce on their local model. FHN ensures that different demographic …
-
Reducing Manual Intervention in Deep Learning
… we introduce Multi-Rate VAE, which uses hypernetworks to learn optimal parameters across all KL weight values in a single training run, eliminating the need for repeated experiments to explore the rate-distortion tradeoff. For optimization, we present Lookahead, an algorithm that …
-
Deep Learning for Data-scarce Environments
… when data is scarce. We propose Multi-Task Hypernetworks, a multi-task learning architecture that improves the modelling of data-scarce environments in two ways. Our approach leverages task-level information explicitly as a supplementary knowledge source to learn task relationships. This …
-
Repurposing large pretrained diffusion models for unsupervised visual understanding and efficient adaptation
… a single image. We leverage this by training hypernetworks to quickly estimate network weights for subject personalized generation, whose convergence is only possible due to the smooth underlying representation of concepts learned by these models. This PhD thesis leverages large pretrained …
-
Applications of Complexity Theory in an English Metropolitan Police Force
… of simulated policy conditions. The nature of hypernetworks introduce non-linearity into policy design, as multiple dimensions are being considered to achieve objectives, where combinations are not predictable from individual dimensions. A central tenet of the thesis is complexity science can …
-
Assessing and Improving Generalization in Graph Reasoning and Learning
… our task, we develop a model based on Graph HyperNetworks and train it on our dataset of neural architecture graphs. Our model can predict performant parameters for unseen deep networks, such as ResNet-50, in a single forward pass. Our model is useful for neural architecture search and …
-
Physics Meets Data: Merging Physics-Based Methods with Deep Learning to Model Complex Systems
… for cardiac modeling is developed that leverages hypernetworks to generate patient-specific models, enabling fast personalization without retraining. Collectively, these contributions advance hybrid modeling by unifying physics-based and data-driven approaches, resulting in robust, interpretable …
-
Natural Language Processing for Healthcare: Text Representation, Multitask Learning, and Applications
… used to generate task-specific parameters with hypernetworks. Experiments show that the proposed hypernetwork-guided multitask learning method can achieve better predictive performance and semantic task information can improve the generalizability of the task-conditioned multitask model.
-
Multi-task generalization for robotics
… decoupling via the hierarchical architecture of Hypernetworks (HNs), networks that generate the parameters of a base network, and investigate how to successfully train HNs via policy distillation. HyperDistill, our proposed method that combines these two key components, achieves similar …
-
Σειριοποίηση προϊόντων πλατφόρμας υπό την επίδραση λειτουργικών περιορισμών
… options of management, was performed using hypernetworks.
Page 1 of 2