Domain Adaptation with Conditional Transferable Components
Authors: Mingming Gong, Kun Zhang, Tongliang Liu, Dacheng Tao, Clark Glymour, Bernhard Schölkopf
ICML 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section we present experimental results on both simulated and real data to show the effectiveness of the proposed CIC and CTC method. |
| Researcher Affiliation | Academia | 1 Centre for Quantum Computation and Intelligent Systems, FEIT, University of Technology Sydney, NSW, Australia 2 Department of Philosophy, Carnegie Mellon University, Pittsburgh, USA 3 Max Plank Institute for Intelligent Systems, T ubingen 72076, Germany |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any concrete access information (e.g., repository link, explicit release statement, or mention of code in supplementary materials) for the source code of the described methodology. |
| Open Datasets | Yes | We also compare our approaches with alternatives on the Office-Caltech dataset introduced in (Gong et al., 2012). The Office-Caltech dataset was constructed by extracting the 10 categories common to the Office dataset (Saenko et al., 2010) and the Caltech256 (Griffin et al., 2007) dataset. We use the bag of visual words features provided by (Gong et al., 2013) for our evaluation. |
| Dataset Splits | Yes | The regularization parameter C of SVM are selected by 5-fold cross validation on a grid. |
| Hardware Specification | No | The paper does not provide specific details regarding the hardware used for running the experiments (e.g., CPU, GPU models, or cloud computing specifications). |
| Software Dependencies | No | The paper does not provide specific ancillary software details, such as library names with version numbers, needed to replicate the experiment. |
| Experiment Setup | Yes | For Gaussian kernel used in MMD, we set the standard deviation parameter σ to the median distance between all source examples. The regularization parameters of the LS transformation are set to λS = 0.001 and λL = 0.0001. ... The regularization parameter for the target information preserving (TIP) term is set to λ = 0.001... We use β-weighted support vector machine (SVM) and weighted kernel ridge regression (KRR)... We use linear kernel for simulation data and Gaussian kernel for real data. The regularization parameter C of SVM are selected by 5-fold cross validation on a grid. |