Fast Prediction for Large-Scale Kernel Machines
Authors: Cho-Jui Hsieh, Si Si, Inderjit S Dhillon
NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We apply our algorithm to real world large-scale classification and regression datasets, and show that the proposed algorithm is consistently and significantly better than other competitors. For example, on the Covertype classification problem, in terms of prediction time, our algorithm achieves more than 10000 times speedup over the full kernel SVM, and a two-fold speedup over the state-of-the-art LDKL approach , while obtaining much higher prediction accuracy than LDKL (95.2% vs. 89.53%). |
| Researcher Affiliation | Academia | Cho-Jui Hsieh, Si Si, and Inderjit S. Dhillon Department of Computer Science University of Texas at Austin Austin, TX 78712 USA {cjhsieh,ssi,inderjit}@cs.utexas.edu |
| Pseudocode | Yes | Algorithm 1: Kernel Approximation with Pseudo Landmark Points. Our overall algorithm DC-Pred++ is presented in Algorithm 2. Algorithm 2: DC-Pred++: our proposed divide-and-conquer approach for fast Prediction. |
| Open Source Code | No | The paper does not provide any specific links to source code repositories or explicitly state that the source code for their methodology is available. |
| Open Datasets | Yes | We use six public datasets (shown in Table 1) for the comparison of kernel SVM prediction time. We further demonstrate the benefits of DC-Pred++ for fast prediction in kernel ridge regression problem on five public datasets listed in Table 2. |
| Dataset Splits | Yes | The parameters γ, C are selected by cross validation, and the detailed description of parameters for other competitors are shown in Appendix 7.1. The parameters used are chosen by five fold cross-validation (see Appendix 7.1). |
| Hardware Specification | Yes | All the experiments are conducted on a machine with an Intel 2.83GHz CPU with 32G RAM. |
| Software Dependencies | No | The paper mentions solving linear SVM problems using LIBLINEAR [6], but it does not specify a version number for this software or any other key software components used in the experiments. |
| Experiment Setup | Yes | The parameters γ, C are selected by cross validation, and the detailed description of parameters for other competitors are shown in Appendix 7.1. The parameters used are chosen by five fold cross-validation (see Appendix 7.1). To control the prediction cost, for Nys, KNys, and DC-Pred++, we vary the number of landmark points, and for RKS and fastfood, we vary the number of random features. |