Learning Mahalanobis Distance Metric: Considering Instance Disturbance Helps
Authors: Han-Jia Ye, De-Chuan Zhan, Xue-Min Si, Yuan Jiang
IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, empirical investigations are conducted to validate the effectiveness of DRIFT. In detail, we first show the interpretability of the process of DRIFT on synthetic data, then DRIFT is compared with state-of-the-art methods on the real datasets. At last, we demonstrate the robustness of DRIFT given perturbed side information and instances. |
| Researcher Affiliation | Academia | National Key Laboratory for Novel Software Technology Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing University, Nanjing, 210023, China {yehj, zhandc, sixm, jiangy}@lamda.nju.edu.cn |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statements about open-sourcing the code for the described methodology or links to code repositories. |
| Open Datasets | Yes | We compare the proposed DRIFT with state-of-the-art metric learning methods on 15 real datasets over 30 random trials. ... Table 1: Comparisons of classification performance (test errors, mean std.) based on 3NN. DRIFTB and DRIFTS are compared. The best performance on each dataset is in bold. Last two rows list the Win/Tie/Lose counts of DRIFTB/S against other methods on all datasets with t-test at significance level 95%. |
| Dataset Splits | No | In each trial, 70% of training data is randomly selected, and the rest is used for test. Parameters are tuned for each method ranging from {10 2, 10 1, . . . , 102}. The paper describes a 70/30 train/test split but does not explicitly mention a separate validation set split or how hyperparameter tuning was performed in relation to a specific validation set. |
| Hardware Specification | No | The paper does not provide any specific hardware details used for running the experiments. |
| Software Dependencies | No | The paper does not specify any software dependencies with version numbers. |
| Experiment Setup | Yes | Parameters are tuned for each method ranging from {10 2, 10 1, . . . , 102}. In the implementation, we initialize metric M = I and α as zero vector. Triplets are initialized the same way as LMNN. |