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 7 of 7 for “"grokking"”.
-
Grokking Applied to Chaotic Iterates of the Logistic Map
This thesis investigates grokking, the delayed transition from memorization to generalization in neural networks trained on deterministic chaotic data. Using an integer–arithmetic discretization of the logistic map, yn+1 =( a yn(p − yn))/ p 2 , bounded aperiodic sequences were generated across …
-
Blurring boundaries : towards the collective team grokking of software product requirements
… manage to achieve varying degrees of collective grokking of the product domain. However, little is known about what factors support or impede these teams in collectively achieving this deep understanding. Looking to identify factors that would explain why some teams collectively grok the product …
-
Analysis of Neural Networks with Physics Applications
… the thesis focuses on the phenomenon of grokking, a delayed generalization effect where models initially overfit but eventually learn to generalize well after extended training. Through a series of interconnected studies, this work proposes insights and practical tools to diagnose, …
-
GENERALIZATION AND TRUSTWORTHINESS IN DEEP LEARNING THROUGH THE LENS OF INFORMATION-THEORETIC MEASURES
… and interesting phenomena in DNNs such as grokking. For application to trustworthy machine learning, we first introduce a pointwise variant of SMI and compare it with Pointwise MI and VI for generating confidence scores and saliency maps in DNNs. Lastly, we apply VI to analyze adversarially …
-
Inductive bias from layerwise structure and backpropagation in neural networks: analysis through simplified models and empirical frameworks
… scaling laws, emergence, lazy/rich regimes, and grokking observed in practical models. Finally, we revisit the long-standing question of how dynamics correlate with performance, offering insights for future research.</p> <p>Because inductive biases are often intractable or ill-defined, we begin …
-
Science of Deep Learning: From Initialization to Emergent Structures
… we demonstrate that emergent abilities such as grokking and in-context learning arise alongside the formation of interpretable structures within the model’s parameters, hidden representations, and outputs. Through targeted experiments, we identify these structures using (i) black-box probing, …