Proximity Preserving Binary Code Using Signed Graph-Cut

Authors: Inbal Lavi, Shai Avidan, Yoram Singer, Yacov Hel-Or4535-4544

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate Proximity Preserving Code on several public datasets: MNIST (Deng 2012), CIFAR-10 (Krizhevsky, Nair, and Hinton 2014), and Label Me (Russell et al. 2008).
Researcher Affiliation Academia 1Department of Electrical Engineering, Tel-Aviv University, Israel 2Department of Computer Science, Princeton University, NJ, USA 3School of Computer Science, The Interdisciplinary Center, Israel
Pseudocode Yes Algorithm 1 Vector Update (b, W) and Algorithm 2 Bit Update (b, W)
Open Source Code No The paper does not provide an explicit statement or link for the open-sourcing of their code.
Open Datasets Yes We evaluate Proximity Preserving Code on several public datasets: MNIST (Deng 2012), CIFAR-10 (Krizhevsky, Nair, and Hinton 2014), and Label Me (Russell et al. 2008).
Dataset Splits No CIFAR-10 ... It is split into 59,000 images in the training set and 1000 in the test set. MNIST ... The dataset is split into a training set of 69,000 samples and a test set of 1,000 samples. No explicit mention of a validation set or split for these datasets.
Hardware Specification No No specific hardware details (e.g., GPU/CPU models, memory) are provided.
Software Dependencies No We use kernel SVM (Scholkopf and Smola 2001) with Gaussian kernels to classify the points {xi} into 1, but any standard classifier can be applied similarly. No specific version numbers for SVM library or any other software components are provided.
Experiment Setup No No specific hyperparameters (e.g., learning rate, batch size, epochs, optimizer settings) or detailed training configurations are provided for the proposed PPC method.