会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 58. 发明公开
    • Cache memory
    • 高速缓存存储器
    • EP0412245A3
    • 1992-01-29
    • EP90108527.4
    • 1990-05-07
    • International Business Machines Corporation
    • Robinson, John Timothy
    • G06F12/12
    • G06F12/124G06F12/122
    • Methods and apparatus are set forth for use with a cache memory resource (that includes a plurality of cache blocks for storing data) and a cache directory (for keeping track of which of said blocks are in use, the number of times each block is referenced and block age), for determining which of said plurality of cache blocks is to be replaced with data to be stored in cache memory on a cache miss. The methods and apparatus cause a reference count, for each cache block, to be maintained in the cache directory; utilize at least one preselected age boundary threshold to determine when to adjust a reference count for a given block on a cache hit; and select a cache block for replacement as a function of reference count value and block age. For example, when 2 preselected age boundary thresholds are utilized, blocks may be considered for replacement only if they have aged past the second boundary and have a reference count below a preselected threshold value. On a cache miss, the block to be replaced can be selected from the set of blocks whose reference counts are below a preselected reference count threshold value thereby allowing for the possibility of non least recently used block replacement choices. Furthermore, according to another embodiment of the invention, to facilitate rapid identification of possible block replacement choices, a chain of blocks is maintained for each reference count below a preselected reference count threshold value.
    • 阐述了用于高速缓存存储器资源(其包括用于存储数据的多个高速缓存块)和高速缓存目录(用于跟踪正在使用哪个所述块,每个块被参考的次数 和块时限),用于确定所述多个高速缓存块中的哪一个将被高速缓存未命中将要存储在高速缓存存储器中的数据替换。 这些方法和装置为每个高速缓存块引起一个引用计数保持在高速缓存目录中; 利用至少一个预选年龄界限阈值来确定何时调整高速缓存命中上给定块的参考计数; 并根据引用计数值和块的年龄选择一个用于替换的高速缓存块。 例如,当使用2个预先选择的年龄边界阈值时,只有当它们已经超过第二边界并且具有低于预选阈值的参考计数时,块才可以被考虑用于替换。 在高速缓存未命中时,可从其参考计数低于预选参考计数阈值的一组块中选择要被替换的块,从而考虑到最近使用的块替换选择的可能性。 此外,根据本发明的另一实施例,为了便于快速识别可能的块替换选择,对于低于预选参考计数阈值的每个参考计数保持块链。
    • 59. 发明公开
    • 결정 기계용 특징 축소 방법
    • 决策机的特征减少方法
    • KR1020070092727A
    • 2007-09-13
    • KR1020077015474
    • 2005-12-14
    • 더 유니버시티 오브 퀸스랜드
    • 게이츠,케빈,이.
    • G06F9/44G06F9/45
    • G06N99/005G06K9/6232G06F12/124G06F15/8053G06F17/30324
    • A method for feature reduction in a training set for a learning machine such as a Support Vector Machine (SVM). In one embodiment the method includes a step (35) of receiving input training data vectors x i of a training set. The input training data vectors are typically derived from a set of features in a feature space. At step (37) the input data vectors are mapped into a multi-dimensional space. At step (39) a least squares problem, derived from a formulation of the SVM, is solved to determine which features comprising the training vectors are to be deemed significant. At step (41) decision parameters and vectors of the chosen decision machine, e.g. SVM, are determined using the features determined to be significant in step (39).
    • 一种用于诸如支持向量机(SVM)的学习机器的训练集中的特征减少的方法。 在一个实施例中,该方法包括接收训练集的输入训练数据矢量x i的步骤(35)。 输入训练数据向量通常从特征空间中的一组特征导出。 在步骤(37),将输入数据矢量映射成多维空间。 在步骤(39),解决从SVM的公式导出的最小二乘问题,以确定包括训练向量的哪些特征被认为是重要的。 在步骤(41),所选决策机的决策参数和向量,例如, SVM使用在步骤(39)中确定为显着的特征来确定。