Discrete Graph Hashing
Authors: Wei Liu, Cun Mu, Sanjiv Kumar, Shih-Fu Chang
NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments performed on four large datasets with up to one million samples show that our discrete optimization based graph hashing method obtains superior search accuracy over state-of-the-art unsupervised hashing methods, especially for longer codes. |
| Researcher Affiliation | Collaboration | Wei Liu Cun Mu Sanjiv Kumar Shih-Fu Chang IBM T. J. Watson Research Center Columbia University Google Research weiliu@us.ibm.com cm3052@columbia.edu sfchang@ee.columbia.edu sanjivk@google.com |
| Pseudocode | Yes | Algorithm 1 Signed Gradient Method (SGM) for B-Subproblem and Algorithm 2 Discrete Graph Hashing (DGH) |
| Open Source Code | No | The paper mentions using publicly available code for competing methods but does not provide concrete access or an explicit statement about the availability of its own source code for the methodology described. |
| Open Datasets | Yes | We conduct large-scale similarity search experiments on four benchmark datasets: CIFAR-10 [15], SUN397 [40], You Tube Faces [39], and Tiny-1M. |
| Dataset Splits | No | The paper specifies how the test sets are formed for each dataset (e.g., '100 images are sampled uniformly randomly from each object category to form a separate test (query) set of 1K images' for CIFAR-10), but does not explicitly detail a validation split or its purpose. |
| Hardware Specification | No | The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers) needed to replicate the experiment. |
| Experiment Setup | Yes | For DGH-I and DGH-R, we set the penalty parameter ρ to the same value in [0.1, 5] on each dataset, and fix TR = 100, TB = 300, TG = 20. IMH, 1-AGH, 2-AGH, DGH-I and DGH-R also use m = 300 anchors (obtained by K-means clustering with 5 iterations)... we adopt the same construction parameters s, t on each dataset (s = 3 and t is tuned following AGH), and ℓ2 distance as D( ). |