Tony's Log

Algorithms, Distributed System, Machine Learning
  博客园 :: 首页 :: 博问 :: 闪存 :: 新随笔 :: 联系 :: 订阅 订阅 :: 管理 ::

随笔分类 -  Thoughts

LeetCode Potential Thought Pitfalls
摘要：Problem Reason Reference Moving ZeroesSort Colors Corner cases Shortest Word Distance Thought: 2 ptr chasing Merge 2 sorted ListsRemove Linked List El 阅读全文
posted @ 2016-05-18 09:34 Tonix 阅读(127) 评论(0) 推荐(0)

Hey... I have different view on training algorithm now
摘要：Basically speaking, I got higher resolution view into code now, in a 庖丁解牛 style. Previously, I treated one program as a whole, and sometimes I got con... 阅读全文
posted @ 2015-05-15 03:28 Tonix 阅读(116) 评论(0) 推荐(0)

Dancing Links - nuclear weapon for Exact Cover problems
摘要：http://www.cnblogs.com/grenet/p/3145800.htmlBest elaboration on dancing list I found. in Chinese.Traditional backtracing is a heuristic-like procedure... 阅读全文
posted @ 2015-03-05 07:38 Tonix 阅读(159) 评论(0) 推荐(0)

Careercup\Mitbbs
摘要：- G: a vector of strings, find a pair with max of strlen(a) * strlen(b) 1. O(n*k): get std::bitset() of each string 2. O(nlgn): sort strings by leng... 阅读全文
posted @ 2015-01-18 11:22 Tonix 阅读(172) 评论(0) 推荐(0)

Feedback from Baozi
摘要：[Technical]1. difference btw. realloc() and free(): realloc() simply changes the size of a mem2. More about TCP\IP, its flow control[Behavior]1. be mo... 阅读全文
posted @ 2015-01-15 14:15 Tonix 阅读(89) 评论(0) 推荐(0)

Beyond ACM - Nuclear-level weapons of data structure
摘要：http://www.zhihu.com/question/26547156Tango Tree Taking advantage of cache locality. 'Preferred path' is a recently access searching path (which is w... 阅读全文
posted @ 2014-11-13 15:50 Tonix 阅读(178) 评论(0) 推荐(0)

TODO List
摘要：1. LC\EPI summary Quick Sort: in C and in Haskell2. G's slides NP-Complete: salesman\knapsack(encrypted) http://blog.csdn.net/liyangbing315/artic... 阅读全文
posted @ 2014-11-01 01:11 Tonix 阅读(197) 评论(0) 推荐(0)

Algorithm & Design
摘要：[Rush]a. k-sum generalized-------------------------!1. Greedy vs. DP:http://blog.csdn.net/yelbosh/article/details/7649717. Multiple choices or not:Gr... 阅读全文
posted @ 2014-11-01 01:09 Tonix 阅读(199) 评论(0) 推荐(0)

懒
摘要：找到自己切题的主要障碍了：试图刚读完题就foresee到最后的solution，其实就是妄图利用比较省事的感性直接超越逻辑推导的理性：其实就是懒。超验感性实际上是建立在大量理性的思考与实践上的。这话不是我说的，出自一位顶尖数学家之口。更别说偶了。 阅读全文
posted @ 2014-08-19 12:40 Tonix 阅读(149) 评论(0) 推荐(0)

Data Structures
摘要：-!5. Trie\Suffix Tree:http://blog.csdn.net/v_july_v/article/details/6897097 Trie: think about how you used King Dict by typing; Suffix Tree: all suff... 阅读全文
posted @ 2014-08-16 14:06 Tonix 阅读(213) 评论(0) 推荐(0)

Some thoughts..
摘要：I think it is natural for an algorithm rookie that when facing a challenging problem, a rookie may fall into her\his algorithm dictionary - she focuse... 阅读全文
posted @ 2014-08-06 13:51 Tonix 阅读(152) 评论(0) 推荐(0)

Some thoughts on a progress
摘要：I can feel that I am making great progress now.. if inspected closely, it is obvious that what I'm getting through is, a newbie phase:coding details m... 阅读全文
posted @ 2014-06-24 14:29 Tonix 阅读(109) 评论(0) 推荐(0)