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 87 for “"WORLD MODELS"”.

  1. World models and planning

    <p>World modeling Is a rather new approach to forecasting the future trends of the complex systems that exist in our world. Therefore, it has a great deal to offer to policy-makers and planners in planning for the future. However, two problems exist with world models and their use in the planning …

    ut-knoxville Repository record for World models and planning (opens in a new tab)

  2. Embodied Intelligence Through World Models

    … and increase living standards around the world. Despite remarkable progress, today's learning algorithms require a large number of examples to succeed at prediction, generation, or control tasks. Unlike these algorithms, humans accumulate common sense knowledge by building internal models …

    toronto-retro Repository record for Embodied Intelligence Through World Models (opens in a new tab)

  3. Neural world models supporting flexible behaviour

    … to flexibly adapt to changes in the external world or internal goals. In order to achieve such flexible behaviour, it has been argued that humans and animals maintain sophisticated internal models – or cognitive maps – of the environment that monitor the state of the world and encode the …

    oxford Repository record for Neural world models supporting flexible behaviour (opens in a new tab)

  4. Interactive multi-agent world models for embodied autonomy

    … and failures are inherently interactive. Real-world testing is prohibitively expensive, necessitating realistic and interactive simulation frameworks. This thesis builds, controls, and reasons through Interactive Multi-Agent World Models for autonomous driving. First, the thesis addresses …

    texas Repository record for Interactive multi-agent world models for embodied autonomy (opens in a new tab)

  5. Causal inductive biases for learning Structured World Models

    World Models have emerged as a powerful paradigm for building autonomous agents: by learning a predictive model of the environment, agents can plan and make decisions based on imagined future scenarios. Despite their success, existing approaches to learning world models largely rely on dense, …

    oxford Repository record for Causal inductive biases for learning Structured World Models (opens in a new tab)

  6. Robot skill adaptation and mastery with world models

    … comprehensive understanding of their surrounding world and its dynamics, limiting their ability to learn subsequent behaviors more efficiently.<br><br>In this thesis, we propose multiple techniques to address the challenges of enabling robots to learn about and adapt to unstructured environments. …

    freiburg-diss Repository record for Robot skill adaptation and mastery with world models (opens in a new tab)

  7. End-to-end Autonomy with Spatial-Temporal World Models

    … (AI) have started to interact with our physical world to carry out a variety of tasks. To make the broad deployment of this technology successful to safety-critical applications such as self-driving vehicles, these embodied AI agents must achieve a stellar understanding of the world we live in. …

    toronto-retro Repository record for End-to-end Autonomy with Spatial-Temporal World Models (opens in a new tab)

  8. Learning to Reach Goals from Suboptimal Demonstrations via World Models

    … to learn faster. In this thesis, we propose World Model Contrastive Reinforcement Learning (WM-CRL), which augments CRL with representations from a world model (WM). The world model is trained to anticipate future state embeddings from past state–action pairs, thereby encoding the dynamics of …

    waterloo Repository record for Learning to Reach Goals from Suboptimal Demonstrations via World Models (opens in a new tab)

  9. LEARNING TO ACT FROM DIVERSE DATA SOURCES VIA WORLD MODELS

    … learn generalist agents? I argue that learning world models that predict future outcomes of actions directly from image observations is a uniquely suitable approach for training generalizable agents. I will discuss how world models can enable powerful unsupervised exploration and how to use a …

    penn Repository record for LEARNING TO ACT FROM DIVERSE DATA SOURCES VIA WORLD MODELS (opens in a new tab)

  10. Learning to model what matters – Representations and world models for efficient reinforcement learning

    … problem can be mitigated in observation-space models by using the value function gradient to measure its sensitivity with regard to model errors. We then move on to combine our insights into representation learning and model learning. We discuss the family of value-aware model learning …

    toronto-retro Repository record for Learning to model what matters – Representations and world models for efficient reinforcement learning (opens in a new tab)

  11. A Deep Hierarchical Variational Autoencoder for World Models in Complex Reinforcement Learning Environments

    … learning (MBRL) approaches leverage learned models of the environment to plan and make optimal decisions, reducing the need for extensive real-world interactions and enabling more efficient learning in complex domains such as robotics, autonomous systems, and resource allocation problems. …

    portland-state Repository record for A Deep Hierarchical Variational Autoencoder for World Models in Complex Reinforcement Learning Environments (opens in a new tab)

  12. Five-dimensional gravitational brane-world models: solutions for black strings and black holes

    … in the framework of five-dimensional braneworld models. The main motivation for studying such theories stems from the Randall-Sundrum model, which was formulated by Lisa Randall and Raman Sundrum back in 1999. To be more precise, we investigate thoroughly the physical characteristics of …

    greece Repository record for Five-dimensional gravitational brane-world models: solutions for black strings and black holes (opens in a new tab)

  13. Artificial General Intelligence (AGI)-Native Wireless Systems: Digital Twins and World Models for Beyond 6G Networks

    … that perform poorly in non-stationary, real-world environments. On the other hand, physical AI agents still lack the capability to generalize and adapt in unforeseen scenarios that appear in real-world settings. As a result, today's AI-native wireless systems and, in turn, their physical AI …

    vt Repository record for Artificial General Intelligence (AGI)-Native Wireless Systems: Digital Twins and World Models for Beyond 6G Networks (opens in a new tab)

  14. Transforming to an age friendly world : models of co-design for creating age friendly public green spaces in urban neighbourhoods

    … City of Unley, explored innovative participatory models by involving older adults as co-designers of age-friendly neighbourhood public green spaces within the urban context. Using citizen science in Stage 1, and then 2D plans, AR & VR in Stage 2, older residents re-designed a local park. Findings …

    adelaide Repository record for Transforming to an age friendly world : models of co-design for creating age friendly public green spaces in urban neighbourhoods (opens in a new tab)

  15. Exploration-Exploitation Trade-Off in Reinforcement Learning

    … of reinforcement learning in large language models and world models have further underscored the critical need for principled, scalable, and efficient exploration algorithms. Exploration has been well-studied in small-scale, model-based reinforcement learning, particularly through the popular …

    tamu Repository record for Exploration-Exploitation Trade-Off in Reinforcement Learning (opens in a new tab)

  16. Exploration-Exploitation Trade-Off in Reinforcement Learning

    … of reinforcement learning in large language models and world models have further underscored the critical need for principled, scalable, and efficient exploration algorithms. Exploration has been well-studied in small-scale, model-based reinforcement learning, particularly through the popular …

    tamu Repository record for Exploration-Exploitation Trade-Off in Reinforcement Learning (opens in a new tab)

  17. World modeling and black-box optimization with language models

    … prior knowledge into their internal world models. It further investigates the use of large language models (LLMs) for black-box optimization (BBO), where an unknown objective must be optimized through trial-and-error, with applications such as prompt optimization and symbolic …

    aalto Repository record for World modeling and black-box optimization with language models (opens in a new tab)

  18. Learning Structured Representations of the Visual World

    Humans develop complex internal models of the world which allow us to generalize remarkably well to new scenarios and tasks. While deep learning has steadily improved in performance through data and scale, it conspicuously lags behind in its generalization to changing data distributions and …

    washington Repository record for Learning Structured Representations of the Visual World (opens in a new tab)

Page 1 of 5