资 源 简 介
该算法在某个再生核希尔伯特空间,在充分考虑领域间分布的均值差和散度差最小化的基础上,基于结构风险最小化模型,提出一种领域适应核支持向量学习机(Kernel support vector machine for domain adaptation, DAKSVM)及其最小平方范式,通过寻求某个特征变换,使得在确保训练数据的最大分割的同时,实现领域间不同分布之间的距离充分最小化,从而实现领域适应学习。
文 件 列 表
EmailSpam
calc_kernel_S.m
calc_mmd.m
compute_beta.m
compute_sv.m
data
load_data.m
load_User00_vs_User01.m
load_User01_vs_User02.m
load_User02_vs_User00.m
main_amkl.m
main_asvm.m
main_cd_svm.m
main_dtmkl_f.m
main_kmm.m
main_mkl.m
main_svm_a.m
main_svm_at.m
main_svm_fr.m
main_svm_t.m
README
return_kernel.m
return_kernel_dtmkl_f.m
run_a_mkl.m
run_a_svm.m
run_cd_svm.m
run_dtmkl_f.m
run_exp_User00_vs_User01.m
run_exp_User01_vs_User02.m
run_exp_User02_vs_User00.m
run_kmm.m
run_mkl.m
run_svm_a.m
run_svm_at.m
run_svm_fr.m
run_svm_t.m
setpaths.m
show_result_a_mkl.m
show_result_a_svm.m
show_result_cd_svm.m
show_result_dtmkl_f.m
show_result_kmm.m
show_result_mkl.m
show_result_svm_a.m
show_result_svm_at.m
show_result_svm_fr.m
show_result_svm_t.m
train_amkl
train_dtmkl_f.m
utils