دانشجو علیرضا صدیقی مقدم دانشجوی کارشناسی ارشد آقای دکتر محمدرضا محمدی
مورخ : ۱۴۰۴/۰۲/۳۰ ساعت۱۴:۰۰ از پروژه کارشناسی ارشد خود با عنوان " پایان نامه جهت دریافت درجه کارشناسی ارشد در رشته مهندسی کامپیوتر – هوش مصنوعی و رباتیکز"دفاع خواهند نمود.
|
ارائه دهنده:
علیرضا صدیقی مقدم
استاد راهنما:
دکتر محمدرضا محمدی
هیات داوران:
ستاد داور داخلی: دکتر محسن سریانی
استاد داور خارجی: دکتر بهروز نصیحت کن
تاریخ دفاع: ۱۴۰۴/۰۲/۳۰
زمان:
۱۴:۰۰
مکان: دانشکده مهندسی کامپیوتر، طبقه سوم ،اتاق سمینار کارشناسی ارشد
|
|
Abstract:
Labeled data is a fundamental component in training supervised deep learning models for computer vision tasks. However, the labeling process, especially for ordinal image classification problems where class boundaries are often ambiguous, is prone to error and noise. Such label noise can sig- nificantly degrade the performance and reliability of machine learning models. This thesis addresses the problem of detecting and correcting label noise in ordinal image classification tasks.
To this end, a novel data-centric method called ORDinal Adaptive Correction (ORDAC) is proposed for adaptive correction of noisy labels. The proposed approach leverages the capabilities of Label Distribution Learning (LDL) to model the inherent ambiguity and uncertainty present in ordinal la- bels. During training, ORDAC dynamically adjusts the mean and standard deviation of the label distribution for each sample. Rather than discarding potentially noisy samples, this approach aims to correct them and make optimal use of the entire training dataset.
The effectiveness of the proposed method is evaluated on benchmark datasets for age estimation (Adience) and disease severity detection (Diabetic Retinopathy) under various asymmetric Gaussian noise scenarios. Results show that ORDAC and its extended versions (ORDACC and ORDACR) lead to significant improvements in model performance. For instance, on the Adience dataset with
۴۰% noise, ORDACR reduced the Mean Absolute Error from ۰.۸۶ to ۰.۶۲ and increased the feed-
back retrieval metric from ۰.۳۷ to ۰.۴۹. The method also demonstrated its effectiveness in cor- recting intrinsic noise present in the original datasets. This research indicates that adaptive label correction using label distributions is an effective strategy to enhance the robustness and accuracy of ordinal classification models in the presence of noisy data.
Keywords: Label Error Detection, Noisy Label Correction, Ordinal Classification, Label Noise, Label Distribution Learning, Data-centric Artificial Intelligence |