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 30 for “"Model interpretability"”.

  1. Towards an Artificial Neuroscience: Analytics for Language Model Interpretability

    The growing deployment of neural language models demands greater understanding of their internal mechanisms. The goal of this thesis is to make progress on understanding the latent computations within large language models (LLMs) to lay the groundwork for monitoring, controlling, and aligning …

    mit Repository record for Towards an Artificial Neuroscience: Analytics for Language Model Interpretability (opens in a new tab)

  2. Neural-embedded discrete choice models

    … of integrating theory-based discrete choice models (DCM) and data-driven neural networks. How to benefit from the strengths of both is the overarching question. I propose hybrid structures and strategies to flexibly represent taste heterogeneity, reduce potential biases, and improve …

    mit Repository record for Neural-embedded discrete choice models (opens in a new tab)

  3. A novel spatiotemporal framework for efficient traffic prediction and visualization

    … Linear Regression-based traffic prediction model, and an interactive map-based traffic simulator to visualize the results. To collect traffic data, we have developed an open-source web-based data scraper tool to extract and export publicly available traffic data from the Google Maps web …

    uoit Repository record for A novel spatiotemporal framework for efficient traffic prediction and visualization (opens in a new tab)

  4. Software defect localization using explainable deep learning

    … software vulnerability detection, where models are trained to classify code as either vulnerable or clean. These models offer advantages over traditional static application testing tools, including adaptability to project-specific code and tunable decision boundaries. They have shown …

    tu-berlin Repository record for Software defect localization using explainable deep learning (opens in a new tab)

  5. Detecting food safety risks and human tracking using interpretable machine learning methods/

    … have allowed researchers to design accurate models using large amounts of data at the cost of interpretability. Model interpretability not only improves user buy-in, but in many cases provides users with important information. Especially in the case of the classification problems addressed in …

    mit Repository record for Detecting food safety risks and human tracking using interpretable machine learning methods/ (opens in a new tab)

  6. Deep learning for Alzheimer’s disease: towards the development of an assistive diagnostic tool

    … healthcare workflow. Though machine learning models have strong predictive power, it is challenging to translate a research project into a clinical tool partly due to the lack of a rigorous validation framework. In this dissertation, I presented a range of machine learning models that were …

    bu Repository record for Deep learning for Alzheimer’s disease: towards the development of an assistive diagnostic tool (opens in a new tab)

  7. Learning classification models of cognitive conditions from subtle behaviors in the digital Clock Drawing Test

    … then explored the tradeoffs in performance and interpretability in classifiers built using a number of different subsets of these features and a variety of different machine learning techniques. We used traditional machine learning methods to build prediction models that achieve high accuracy. …

    mit Repository record for Learning classification models of cognitive conditions from subtle behaviors in the digital Clock Drawing Test (opens in a new tab)

  8. Unveiling Phenotype–Genotype Interplay with Deep Learning Foundation Models for scRNA-seq: A Quantitative Perspective

    Foundation models have emerged as powerful tools for analyzing single-cell RNA sequencing (scRNA-seq) data, leveraging large-scale pretraining to capture complex gene expression patterns. However, a comprehensive quantitative framework for understanding the interplay between phenotypes and …

    mit Repository record for Unveiling Phenotype–Genotype Interplay with Deep Learning Foundation Models for scRNA-seq: A Quantitative Perspective (opens in a new tab)

  9. Entropy-based machine learning algorithms applied to genomics and pattern recognition

    … areas for potential improvements for these ML models, including a higher degree of model interpretability and overall accuracy. In this thesis, we present decision tree (DT) methods applied to DNA sequence analysis that result in highly interpretable and accurate predictions. We propose a …

    uiuc Repository record for Entropy-based machine learning algorithms applied to genomics and pattern recognition (opens in a new tab)

  10. Techniques for Interpretability and Transparency of Black-Box Models

    … hinders people's ability to inspect these models. Furthermore, legal requirements are being proposed to require a level of model understanding as a prerequisite to the deployment and use. These factors have spurred research that increases the interpretability and transparency of these …

    mit Repository record for Techniques for Interpretability and Transparency of Black-Box Models (opens in a new tab)

  11. Towards ML Models That We Can Deploy Confidently

    … challenge through two key thrusts: (1) making ML models more trustworthy by leveraging what has been perceived solely as a weakness of ML model—adversarial perturbations, and (2) exploring the underpinnings of reliable ML deployment. Specifically, in the first thrust, we focus on adversarial …

    mit Repository record for Towards ML Models That We Can Deploy Confidently (opens in a new tab)

  12. Interpretability by Design: New Interpretable Machine Learning Models and Methods

    <p>As machine learning models are playing increasingly important roles in many real-life scenarios, interpretability has become a key issue for whether we can trust the predictions made by these models, especially when we are making some high-stakes decisions. Lack of transparency has long been a …

    duke Repository record for Interpretability by Design: New Interpretable Machine Learning Models and Methods (opens in a new tab)

  13. Optimization Methods for Machine Learning under Structural Constraints

    In modern statistical and machine learning models, structural constraints are usually imposed for model interpretability as well as model complexity reduction. In this thesis, we present scalable optimization methods for several large-scale machine learning problems under structural constraints, …

    mit Repository record for Optimization Methods for Machine Learning under Structural Constraints (opens in a new tab)

  14. An Analysis of Neural Rationale Models andInfluence Functions for Interpretable MachineLearning

    … years, increasingly powerful machine learning models have shown remarkable performance on a wide variety of tasks and thus their use is becoming more and more prevalent, including deployment in high stakes settings such as for medical and legal applications. Because these models are complex, …

    mit Repository record for An Analysis of Neural Rationale Models andInfluence Functions for Interpretable MachineLearning (opens in a new tab)

  15. Machine learning-driven integration of multimodal data for deciphering breast cancer heterogeneity

    … images, and clinical outcome data). Besides, the model interpretability and privacy issues should also be carefully taken into consideration in machine learning-based BC research. This thesis aims to explore BC heterogeneity using thriving machine-learning algorithms at different data resolutions …

    manitoba Repository record for Machine learning-driven integration of multimodal data for deciphering breast cancer heterogeneity (opens in a new tab)

  16. When Moneyball Meets the Beautiful Game: A Predictive Analytics Approach to Exploring Key Drivers for Soccer Player Valuation

    … the trade-offs between predictive accuracy and model interpretability. XGBoost, the best model for player valuation, yields the lowest RMSE and the highest adjusted R2. SHAP values identify the most important features in the best model both at a collective level and at an individual level. This …

    brock Repository record for When Moneyball Meets the Beautiful Game: A Predictive Analytics Approach to Exploring Key Drivers for Soccer Player Valuation (opens in a new tab)

  17. Multi-level models of language comprehension in the mind and brain

    … to efficiently construct a lexicon. This modeling work recapitulates classical dynamics of language learning exhibited by children acquiring their first language, and more broadly presents an expanded view of the computational role of these representational systems. The second part of the …

    mit Repository record for Multi-level models of language comprehension in the mind and brain (opens in a new tab)

  18. Next-Generation Intelligent Portfolio Management

    … framework that leverages Transformer-based models and Large Language Models (LLMs) to enhance return predictions and sentiment extraction from extensive financial texts coupled with robust DRL trading agents to optimize portfolio performance. We introduce an adaptive retrieval-augmented …

    mit Repository record for Next-Generation Intelligent Portfolio Management (opens in a new tab)

  19. An Interpretable Multimodal Framework for Regional Organ Transplantation Outcomes

    … thesis, we highlight the use of large language model (LLM) embeddings combined with structured tabular data to build a predictive classifier that estimates offer outcomes for kidney donor-recipient matches. For each predictive model deployed, we provide further analysis on the interpretability

    mit Repository record for An Interpretable Multimodal Framework for Regional Organ Transplantation Outcomes (opens in a new tab)

  20. On internal language representations in deep learning : an analysis of machine translation and speech recognition

    … in this technology thanks to their ability to model large amounts of data. Contrary to traditional systems, models based on deep neural networks (a.k.a. deep learning) can be trained in an end-to-end fashion on input-output pairs, such as a sentence in one language and its translation in …

    mit Repository record for On internal language representations in deep learning : an analysis of machine translation and speech recognition (opens in a new tab)

Page 1 of 2