| _version_ | 1866902096784130048 |
|---|---|
| author | Goudarzi, Hamidreza Heydarian, Ashkan Nasirinasab, Pejman |
| author_facet | Goudarzi, Hamidreza Heydarian, Ashkan Nasirinasab, Pejman |
| contents | <p> This code presents a novel mathematical framework for transfer learning across different<br> modalities in the Internet of Things (IoT) using kernel methods in deep learning. We in<br>vestigate the underlying relationships between various IoT data modalities, such as sensor<br> readings, video feeds, and textual data, and propose techniques that leverage knowledge<br> transfer to enhance model performance on new data distributions. Our approach for<br>malizes these relationships mathematically, enabling effective adaptation even when data<br> originates from disparate IoT sources. We validate our methods through theoretical anal<br>ysis and empirical experiments on IoT datasets. Open app.py by streamline in the same directory with experiment.py</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_17070731 |
| institution | Zenodo |
| language | |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Code of Transfer Learning Across Modalities in IoT: A Mathematical Framework for Kernel-Based Deep Learning Goudarzi, Hamidreza Heydarian, Ashkan Nasirinasab, Pejman <p> This code presents a novel mathematical framework for transfer learning across different<br> modalities in the Internet of Things (IoT) using kernel methods in deep learning. We in<br>vestigate the underlying relationships between various IoT data modalities, such as sensor<br> readings, video feeds, and textual data, and propose techniques that leverage knowledge<br> transfer to enhance model performance on new data distributions. Our approach for<br>malizes these relationships mathematically, enabling effective adaptation even when data<br> originates from disparate IoT sources. We validate our methods through theoretical anal<br>ysis and empirical experiments on IoT datasets. Open app.py by streamline in the same directory with experiment.py</p> |
| title | Code of Transfer Learning Across Modalities in IoT: A Mathematical Framework for Kernel-Based Deep Learning |
| url | https://doi.org/10.5281/zenodo.17070731 |