Discovering a Zero (Zero-Vector Class of Machine Learning)
Authors: Harikrishna Metta, Venkatesh Babu Radhakrishnan
ICML 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Results for the key applications are shown using the MNIST dataset. To further strengthen the claims, some results are also produced using the CIFAR10 and Image Net-1k embeddings. ... The applications discussed in this section are empirically evaluated in the Appendix. Detailed results and analysis can be found in Section E. |
| Researcher Affiliation | Academia | 1Vision and AI Lab (val.cds.iisc.ac.in), Department of Computational and Data Sciences, Indian Institute of Science, Bangalore, INDIA. Correspondence to: Harikrishna Metta <EMAIL>, R. Venkatesh Babu <EMAIL>. |
| Pseudocode | No | The paper describes methods through mathematical equations and textual explanations, but it does not include any clearly labeled 'Pseudocode' or 'Algorithm' blocks or figures that structure procedural steps. |
| Open Source Code | Yes | The code supporting these applications is publicly available at: github.com/hm-4/Metta-Class. |
| Open Datasets | Yes | Results for the key applications are shown using the MNIST dataset. ... To further strengthen the claims, some results are also produced using the CIFAR10 and Image Net-1k embeddings. ... Standard MNIST data set is used for the following results. ... Standard CIFAR10 data set is used for the following results. |
| Dataset Splits | No | The paper mentions dividing data into 'train and test data' and notes that 'The Metta-Class data used to calculate the train accuracy but not used in calculation of test accuracy'. However, it does not specify exact split percentages, sample counts for each split, or reference any particular predefined splits with citations for reproduction. |
| Hardware Specification | No | The paper discusses computational complexity but does not explicitly describe the specific hardware (e.g., GPU models, CPU types, memory) used for running its experiments. |
| Software Dependencies | No | The paper mentions 'Sample Py Torch implementations' but does not provide specific version numbers for PyTorch or any other software libraries or dependencies used in the experiments. |
| Experiment Setup | No | The paper shows plots across 'Epochs' (e.g., Figure 17, 18, 29), and mentions 'initialized with random noise image (x0)' and 'Cross Entropy loss', but it does not provide specific values for hyperparameters such as learning rate, batch size, optimizer settings, or detailed training schedules for the neural networks used. |