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 119 for “"Learning dynamics"”.
-
Deep Learning Dynamics: From Minimization to Games
… witnessed the unprecedented successes of deep learning in various high-profile challenges. Meanwhile, the practical success of deep learning has greatly outpaced our theoretical understanding, and it is tempting to blindly use this hammer without knowing how it works. In particular, the current …
-
Generalization through the lens of learning dynamics
<p>A machine learning (ML) system must learn not only to match the output of a target function on a training set, but also to generalize to novel situations in order to yield accurate predictions at deployment. In most practical applications, the user cannot exhaustively enumerate every possible …
-
Curriculum Learning: from Human Strategies to Learning Dynamics
The recent success of machine learning (ML), or "the third wave of artificial intelligence (AI)", is built upon computational methods from the fields of optimization and statistics, the availability of large-scale training data and computational power, and partial imitation of human cognitive …
-
Social learning dynamics in chimpanzees: Reflections on animal culture
Contains fulltext : 141062.pdf (Publisher’s version ) (Open Access)
-
Older adults’ collaborative learning dynamics when exploring feature-rich software
… through complex structures which can make their learning process difficult. While many studies have investigated software learning, they typically have not focused on older adult users. Given that older adults are increasingly using more complex software, we aimed to investigate how older adults …
-
Learning Dynamics from Data Using Optimal Transport Techniques and Applications
… in computational mathematics, machine learning, engineering, even finance areas. Meanwhile, people are gradually realizing that as the amount of data as well as the needs of utilizing data increase vastly, data-driven models have great potentials in real-world applications. In this …
-
Learning dynamics of deep learning -- force analysis of deep neural networks
This thesis investigates the learning dynamics of deep learning systems through a local, physics-inspired analytical lens. Motivated by the need for fine-grained insights into model behavior, we begin with the step-wise influence that a single training example exerts on a specific observing example …
-
Interpreting Deep Neural Networks and Beyond: Visualization, Learning Dynamics, and Disentanglement
… addition, a better understanding of current deep learning methods will inspire the development of more principled computational approaches with better robustness and interpretability. In this thesis, we focus on interpreting and improving deep neural networks from the following three perspectives: …
-
Reliable and efficient post-training of large reasoning models : reinforcement learning dynamics and adaptation
… reasoning models, with a focus on reinforcement learning (RL) dynamics and efficient adaptation. The first part of the dissertation studies reliable post-training by understanding the learning dynamics of RL. It shows that Group Relative Policy Optimization can suffer from Lazy Likelihood …
-
Towards Efficient Robotic Manipulation of Deformable Objects by Learning Dynamics Models and Adaptive Policies
… reason for this success has been the advent of learning-based approaches, which driven by improvements in deep learning techniques, have endowed robots with greater generalisation capabilities to manipulate objects varying in shape, size, and texture. However, the majority of these …
-
Dynamics of Multi-Agent Learning Under Bounded Rationality: Theory and Empirical Evidence
… development of a principled understanding of the learning dynamics and strategic interactions in human-machine systems. We propose a game-theoretic framework that captures the complexities of multi-agent learning under bounded rationality, focusing on the effects of timescale separation, varying …
-
Essays in Financial Economics
… portfolio optimization, and the economics of learning and belief formation in financial markets. The unifying premise is that a coherent account of risk premia, optimal portfolio choice, and return predictability requires moving beyond the classical mean-variance paradigm to incorporate …
-
Understanding Large-Scale AI: From Training Dynamics to Alignment
… The first part of this thesis focuses on the learning dynamics of large-scale neural networks. I study optimization landscapes and convergence properties of gradient-based methods, with a particular emphasis on the Transformer architecture. By analyzing gradient descent trajectories and …
-
Contests: equilibrium analysis, design and learning
… objectives. In particular, we (i) analyze learning dynamics in Tullock contests using tools from probabilistic analysis of algorithms and optimization, (ii) design contests that improve diversity in participation, and (iii) study the existence, welfare efficiency, and computational …
-
Learning with Less: Low-rank Dynamics, Communication, and Introspection in Machine Learning
… analysis of the optimization methods in machine learning, and the underlying role that the matrix rank of utilized learning statistics plays in these algorithms. We show that this new perspective on machine learning optimization provides benefits in terms of communication-efficient federated …
-
Relational and Analogical Reasoning with Spiking Neural Networks
… reasoning can emerge from general-purpose neural dynamics alone. Domain knowledge is not merely an engineering detail but a conceptual axis shaping model interpretability, generalisation, and flexibility. In many real-world scenarios, the absence of such prior knowledge is precisely what makes …
-
Modeling human-coupled common pool resource systems with techniques in evolutionary game theory and reinforcement learning
… theoretic, and more recently, a reinforcement learning approach. Within game theoretic models, the mechanisms used for learning dynamics are often assumed, and the influence of the type of learning dynamics are not systematically compared under identical models. Models using reinforcement …
-
GENERALIZATION AND TRUSTWORTHINESS IN DEEP LEARNING THROUGH THE LENS OF INFORMATION-THEORETIC MEASURES
… sliced information plane framework to study the learning dynamics 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 …
Page 1 of 6