Discriminative Semi-Supervised Dictionary Learning with Entropy Regularization for Pattern Classification

Authors: Meng Yang, Lin Chen

AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on face recognition, digit recognition and texture classification show the effectiveness of the proposed method.
Researcher Affiliation Academia Meng Yang, Lin Chen 1, 2 1College of Computer Science & Software Engineering, Shenzhen University, Shenzhen, China 2School of Data and Computer Science, Sun Yat-sen University, Guangzhou, China
Pseudocode No The paper describes methods using equations and textual steps but does not include a formally structured pseudocode or algorithm block.
Open Source Code No The paper does not provide an explicit statement or link confirming the availability of its own open-source code for the described methodology.
Open Datasets Yes We evaluate our approach on two face databases: Extended Yale B database (Lee, Jeffrey and David 2005), and LFW face database (Wolf, Hassner and Taigman 2009), two handwritten digit datasets: MNIST (Le Cun et al. 1998) and USPS(Hull 1994) and an object category database: Texture(Lazebnik, Schimid and Ponce 2005) for the tasks of face recognition, digit recognition and texture classification, respectively.
Dataset Splits No The paper specifies training and testing splits, and how training data is further divided into labeled and unlabeled subsets, but does not explicitly mention a separate validation set for hyperparameter tuning or early stopping.
Hardware Specification No The paper does not provide specific hardware details such as GPU/CPU models, memory, or cloud instance types used for running the experiments.
Software Dependencies No The paper does not provide specific software dependencies or library versions (e.g., 'Python 3.8, PyTorch 1.9') needed to replicate the experiment.
Experiment Setup Yes In our all experiments, we set γ=0.001 and λ=0.01 based on our experimental experience. ... we set β=0.01 to suitably lower the weight of the unlabeled samples classified wrongly, while utilize the discrimination of learnt dictionary better.