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 29 for “"Learning Dynamics"”.

  1. Learning dynamics in feedforward neural networks

    Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 1995.

    mit Repository record for Learning dynamics in feedforward neural networks (opens in a new tab)

  2. Learning, dynamics of beliefs, and asset pricing

    … on the observation of price and volume. The learning process of the statistical arbitrageurs leads to an optimal trading strategy that can be upward sloping in prices. The presence of privately informed investors makes the equilibrium price dependent on the history of trading volume. The …

    mit Repository record for Learning, dynamics of beliefs, and asset pricing (opens in a new tab)

  3. 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: …

    rice Repository record for Interpreting Deep Neural Networks and Beyond: Visualization, Learning Dynamics, and Disentanglement (opens in a new tab)

  4. Information and Incentives in Online Platforms

    … in Chapter 2 we examine the outcome of the learning dynamics in traffic routing where the latency functions are unknown. We show that the combination of selfish routing and learning dynamics converges to the full-information Wardrop equilibrium, this supports the study of the Wardrop …

    mit Repository record for Information and Incentives in Online Platforms (opens in a new tab)

  5. Passivity, no-regret, and performance in online learning and games

    … no-regret and performance of continuous-time learning dynamics. The analysis is divided into two parts: (i) the interaction between a learning model and a dynamic, uncertain environment, and (ii) the interaction among multiple adaptive learners within a game. In the first part, the learning

    uiuc Repository record for Passivity, no-regret, and performance in online learning and games (opens in a new tab)

  6. Artificial markets and intelligent agents

    … in a market. A number of features of the market dynamics are examined: the price efficiency of the market, the speed at which prices converge to the rational expectations equilibrium price, and the learning dynamics of traders who possess diverse information or preferences.

    mit Repository record for Artificial markets and intelligent agents (opens in a new tab)

  7. Efficient Learning and Computation of Linear Correlated Equilibrium in General Convex Games

    We propose efficient no-regret learning dynamics and ellipsoid-based methods for computing linear correlated equilibria—a relaxation of correlated equilibria and a strengthening of coarse correlated equilibria—in general convex games. These are games where the number of pure strategies is …

    mit Repository record for Efficient Learning and Computation of Linear Correlated Equilibrium in General Convex Games (opens in a new tab)

  8. Information, Learning and Incentive Design for Urban Transportation Networks

    … (navigation apps) on the strategic behavior and learning processes of travelers in uncertain networks; 2) Market mechanism design for efficient carpooling and toll pricing in the presence of autonomous driving technology; 3) Security analysis and resource allocation for robustness under random or …

    mit Repository record for Information, Learning and Incentive Design for Urban Transportation Networks (opens in a new tab)

  9. Multi-Player Zero-Sum Markov Games with Networked Separable Interactions

    … topology”. Then, we propose fictitious-play-type dynamics, the classical learning dynamics in normal-form games, for zero-sum NMGs, and establish convergence guarantees to Markov stationary NE under a star-shaped network structure. Finally, in light of the hardness result, we focus on computing a …

    mit Repository record for Multi-Player Zero-Sum Markov Games with Networked Separable Interactions (opens in a new tab)

  10. Dynamics of Gradient Flow with Contrastive Learning

    Contrastive learning (CL), in di erent forms, has been shown to learn discriminatory representations for downstream tasks without the need of human labeling. In the representation space learnt via CL, each class collapses to a distinct vertex of a simplex on a hypersphere during training. This …

    mit Repository record for Dynamics of Gradient Flow with Contrastive Learning (opens in a new tab)

  11. Exploring the practice of supporting early childhood students’ numeracy skills through collaborative learning in working group activities

    Collaborative learning has been increasingly recognized as an optimal pedagogical strategy to support early numeracy development. This study explores the application of collaborative learning in early numeracy education at the kindergarten level in Indonesia an area that remains relatively …

    uiii Repository record for Exploring the practice of supporting early childhood students’ numeracy skills through collaborative learning in working group activities (opens in a new tab)

  12. Essays in macroeconomics, corporate finance, and social learning

    … released without delay. This produces asymmetric learning dynamics for the economy, with gradual booms and rapid recessions. In a boom the belief about the aggregate state increases, information is pushed off into the future, and learning is slow. In a recession the belief is falling, triggering a …

    mit Repository record for Essays in macroeconomics, corporate finance, and social learning (opens in a new tab)

  13. No-Regret Learning in General Games

    … investigates the regret performance of no-regret learning algorithms in the competitive, though not fully-adversarial, environment of games. We establish exponential improvements on previously best-known external and internal regret bounds for these settings. We show that Optimistic Hedge – a …

    mit Repository record for No-Regret Learning in General Games (opens in a new tab)

  14. Towards the Efficient, Scientific and Accessible Development of Small Language Models

    … training small models. I investigate curriculum learning strategies informed by human language acquisition in data-constrained settings within a framework called CLIMB, and I introduce Syntactic Smoothing, a cognitively motivated method that enhances the representation of infrequent words by …

    cambridge Repository record for Towards the Efficient, Scientific and Accessible Development of Small Language Models (opens in a new tab)

  15. Die Entwicklung eines sozial-kognitiven Lernmodells zur Verbesserung der Schlüsselkompetenzen von Auszubildenden in der Kraftfahrzeugtechnik in Wien

    … model, grounded in Vygotsky's social-cognitive learning theory. The research methodology employs a dual approach: a quantitative survey of 149 students assessing their current status, acquisition, and enhancement of four pivotal motor vehicle skills, alongside a qualitative exploration through …

    middlesex Repository record for Die Entwicklung eines sozial-kognitiven Lernmodells zur Verbesserung der Schlüsselkompetenzen von Auszubildenden in der Kraftfahrzeugtechnik in Wien (opens in a new tab)

  16. Who is this self I'm supposed to be expressing? : narrative inquiry into the art and learning of twelve women visual arts students

    … perspectives on art making, self concept and learning were revealed. Regardless of age and life experience, these women regard their art learning as catalytic in their growth toward subjectivity, and personal, social and cultural agency. Links are drawn between expectations, life experiences, …

    concordia Repository record for Who is this self I'm supposed to be expressing? : narrative inquiry into the art and learning of twelve women visual arts students (opens in a new tab)

  17. Distributed learning in games under bounded rationality

    … considers models, solution concepts, and learning dynamics that explicitly embrace bounded rationality as a structural feature. Furthermore, we focus on the inherently distributed nature of these systems, where agents make decisions independently based on local information and …

    uiuc Repository record for Distributed learning in games under bounded rationality (opens in a new tab)

  18. Non-equilibrium physics: from spin glasses to machine and neural learning

    … processing tasks such as error correction, learning, and optimization. Despite the empirical success of utilizing these systems for intelligent tasks, the underlying principles that govern their emergent intelligent behaviors remain largely unknown. In this thesis, we aim to characterize …

    mit Repository record for Non-equilibrium physics: from spin glasses to machine and neural learning (opens in a new tab)

  19. Decentralized Machine Learning over Fragmented Data

    … and algorithms to decentralize the machine learning pipeline. Current approaches in this area have centered around Federated Learning (FL), which enables model training over distributed data. However, FL’s dependence on central coordination and inflexibility with heterogeneous systems limit …

    mit Repository record for Decentralized Machine Learning over Fragmented Data (opens in a new tab)

  20. Essays in political economy

    … equilibria, the model shows that limited learning is possible, and that learning dynamics are shaped by an informational commitment problem arising when rulers are "too optimistic". In joint work with James Robinson, the final chapter is a comparative empirical study of the impact of …

    mit Repository record for Essays in political economy (opens in a new tab)

Page 1 of 2