Summary of Informed Deep Hierarchical Classification: a Non-standard Analysis Inspired Approach, by Lorenzo Fiaschi and Marco Cococcioni
Informed deep hierarchical classification: a non-standard analysis inspired approach
by Lorenzo Fiaschi, Marco Cococcioni
First submitted to arxiv on: 25 Sep 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Machine Learning (cs.LG); Logic (math.LO)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed approach to deep hierarchical classification involves designing a novel neural network architecture called lexicographic hybrid deep neural network (LH-DNN). This architecture combines elements from lexicographic multi-objective optimization, non-standard analysis, and deep learning. The LH-DNN is compared to the B-CNN on various benchmarks, including CIFAR10, CIFAR100, and Fashion-MNIST. Results show that LH-DNN can achieve comparable or superior performance with reduced learning parameters, training epochs, and computational time. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper proposes a new way to classify data with multiple labels using a deep neural network called the lexicographic hybrid deep neural network (LH-DNN). This type of classification is important because it helps us understand how things are related. The LH-DNN uses tools from different areas of research, like optimization and non-standard analysis, to help it learn. The paper compares this new approach to an existing method called the B-CNN on several datasets. It shows that the LH-DNN can do just as well or even better than the B-CNN while using fewer resources. |
Keywords
* Artificial intelligence * Classification * Cnn * Deep learning * Neural network * Optimization
