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 101 for “"black box model"”.
-
Disentangling the AI Black-Box Model: From Direct to Indirect Influence
… of artificial intelligence and machine learning models across numerous domains that directly impact individuals’ lives, it is essential to ensure that these models are making their decisions in a non-discriminatory manner. That is, biases encoded in the underlying training data are not reflected …
-
Think outside the Black Box: Model-Agnostic Deep Learning with Domain Knowledge
Deep Learning (DL) models are trained on a downstream task by feeding (potentially preprocessed) input data through a trainable Neural Network (NN) and updating its parameters to minimize the loss function between the predicted and the desired output. While this general framework has mainly …
-
A Black-box Model of Memory Life-span for NAND Flash Devices
… resulting in less overhead. We offer a model to support the new metric with only three parameters compared to the tens of parameters required by the prior art, thus making the scheme suitable for online processing. Experimentally, we achieved a mean accuracy of 3% in predicting the …
-
Program analysis and black-box GUI testing
… to systematically select event sequences uses a black-box model. The black-box model is inferred by recording possible sequences of user interactions from the running GUI program (ignoring the source code of the GUI program). The model informs the tester about the executability of event …
-
Wrapper boxes for increasing model interpretability via example-based explanations
We propose wrapper boxes to provide interpretability in deep learning that is model, training, and dataset-agnostic. The prediction model is trained as usual on some dataset(s), typically optimizing some predetermined loss function. At inference time, the prediction model is augmented by a simpler …
-
Dynamic Data Structures for Geometric Search and Retrieval
… from the past. This is called the <italic>black-box model</italic>. We present efficient online algorithms for maintaining two important geometric data structures for moving points in the black-box model, namely <italic>nets</italic> and <italic>net trees</italic>. We establish the …
-
Article Search Tool and Topic Classifier
… of the words in the article. In the last phase, model-agnostic Explainable AI (XAI) techniques are implemented to prove the findings in this thesis. XAI techniques are also employed to show how we can fine tune model hyper-parameters for a black-box model.
-
Black-box printer models and their applications
… to as dot-gain, is complicated. The printer models which are developed according to a pre-designed test page can either be embedded in the halftoning algorithm, or used to predict the printed halftone image at the input to an algorithm being used to assess print quality. In our research, we …
-
In-training explainability frameworks to make black-box machine learning models more explainable.
Despite ongoing efforts to make black-box machine learning models more explainable, transparent, and trustworthy, there is growing advocacy for using only inherently interpretable models for high-stake decision-making. Post-hoc explanations have recently been criticized for learning surrogate …
-
Intrinsic interpretable machine learning frameworks for image classification
… the predictions performed by intelligent machine models used in practical application domains. Interpretability/explainability in the machine learning field has recently become a significant problem because many real-world applications need justification and explanation for the decisions made by …
-
FORENSIC ANALYSIS AND DEFENSE FOR MACHINE LEARNING MODEL IN BLACK-BOXES
As machine learning models become widely deployed in many areas, there are also rising security concerns. This thesis studies two aspects: forensic analysis and adversarial defenses. In the first half of this thesis, we take the perspective of a forensic investigator who wants to investigate some …
-
STATISTICAL LEARNING FOR STANDARD MODEL PHENOMENOLOGY
… both in terms of parametrization and model selection. We discuss different sources of PDF uncertainty. First, we elucidate the nontrivial aspects of averaging over the space of PDF determinations by explicitly calculating the data-driven correlation between different sets of PDFs. …
-
Identifying and Predicting Rat Behavior Using Neural Networks
… In this thesis, neural networks are used as a black-box model to map electrophysiological data, representative of an ensemble of neurons in the hippocampus, to a T-maze, wheel running or open exploration behavior. The velocity and spatial coordinates of the identified behavior are then …
-
Identifying factors that influence purchasing behaviour of black metrosexual facial skincare products in Gauteng
… behaviour of facial skincare products of black metrosexual men residing in Gauteng. Three seminal consumer behaviour models were used as theoretical base; Howard-Sheth, Engel, Kollat and Miniard, and the Black box model. The study is an exploratory purchasing behavioural study and followed …
-
Impact of Distributed Battery Energy Storage on Electric Power Transmission and Distribution Systems
… grid are studied, for which suitable models for steady-state and dynamic analyses are proposed. Thus, first, a dynamic average BESS model is proposed, which comprises a simplified representation of the battery cells to allow simulating the effects of battery degradation, a …
-
Domain adaptation for image classification and crowd counting
… and crowd counting. Given a pre-trained model learned from a source domain, our goal is to adapt this model to a target domain using unlabeled data. The solution of this problem has a lot of potential applications in computer vision research that require a neural network model adapted to …
-
An Intelligent Nonlinear System Identification Method with an Application to Condition Monitoring
Neural networks are black-box model structures that map inputs to outputs and do not require underlying mathematical models between the two. They are frequently used in the field of system identification, the area that deals with the development of system models based on input-output data. In this …
-
Model-Based Testing of Smart Home Systems Using EFSM, CEFSM, and FSMApp
… and FSMApp [10] to generate reusable test-ready models of smart home systems. We present an approach to create reusable test-ready models of smart home systems using EFSMs to model device components (Sensor, Controller and Actuator), EFSMs to model single devices in the SHS and the interaction …
-
Model-Agnostic Explanations and Evaluation of Machine Learning
… communication with humans: they are functionally black boxes, hard to debug and hard to evaluate properly. This communication is crucial though: humans are the ones who train, deploy and use machine learning models, and thus have to make trust and evaluation decisions. Furthermore, it is humans …
Page 1 of 6