<![CDATA[Mingkui Tan's Homepage - Blog]]>Fri, 08 Apr 2016 07:53:46 -0700Weebly<![CDATA[May 15th, 2014]]>Fri, 16 May 2014 04:02:56 GMThttp://www.tanmingkui.com/blog/may-15th-2014The framework for invariant learning using BVM:

  1. We do tansformation on those fartherest points which should be the supporter of BVM.
  2. We add the new point in the end of prob->x
  3. We allcote the memory in advance by using the param.times optionion. So the number of support vectors couldn’t be larger than param->times*prob->l.
  4. when a new point is added in, we need to update the kernel matrix as well, mainly for the diagonal element calculation.
  5. For the storage, we can allocate more depending on our use. But we should notice that the memory can not be overlapped. 
<![CDATA[May 14th, 2014]]>Thu, 15 May 2014 03:00:05 GMThttp://www.tanmingkui.com/blog/may-14th-2014<![CDATA[What is big data?]]>Thu, 15 May 2014 02:17:57 GMThttp://www.tanmingkui.com/blog/what-is-big-dataBig data has emerged in many applications]]>