没有家庭作业。

1995年，我作为专业程序员的第一份工作是为标准普尔500指数期货编写一个基于遗传算法的自动交易系统。该应用程序是用Visual Basic 3 [!我不知道我当时是怎么做的，因为VB3甚至没有课程。

The application started with a population of randomly-generated fixed-length strings (the "gene" part), each of which corresponded to a specific shape in the minute-by-minute price data of the S&P500 futures, as well as a specific order (buy or sell) and stop-loss and stop-profit amounts. Each string (or "gene") had its profit performance evaluated by a run through 3 years of historical data; whenever the specified "shape" matched the historical data, I assumed the corresponding buy or sell order and evaluated the trade's result. I added the caveat that each gene started with a fixed amount of money and could thus potentially go broke and be removed from the gene pool entirely.

在对种群的每一次评估之后，幸存者被随机杂交(通过混合来自两个亲本的片段)，一个基因被选择为亲本的可能性与它产生的利润成正比。我还添加了点突变的可能性，让事情变得有趣一点。经过几百代这样的基因，我最终得到了一个基因群，它可以把5000美元变成平均约10000美元，而且没有死亡/破碎的可能性(当然是在历史数据上)。

Unfortunately, I never got the chance to use this system live, since my boss lost close to $100,000 in less than 3 months trading the traditional way, and he lost his willingness to continue with the project. In retrospect, I think the system would have made huge profits - not because I was necessarily doing anything right, but because the population of genes that I produced happened to be biased towards buy orders (as opposed to sell orders) by about a 5:1 ratio. And as we know with our 20/20 hindsight, the market went up a bit after 1995.

这是一段时间以前的事了，但我滚动了一个GA来进化实际上是图像处理内核的东西，以从哈勃太空望远镜(HST)图像中去除宇宙射线痕迹。标准的方法是用哈勃望远镜进行多次曝光，只保留所有图像中相同的东西。由于HST时间是如此宝贵，我是一个天文学爱好者，最近参加了进化计算大会，我考虑使用GA来清理单次曝光。

这些个体以树的形式存在，以3x3像素的区域作为输入，执行一些计算，并决定是否以及如何修改中心像素。通过将输出图像与用传统方法(即叠加曝光)清理的图像进行比较来判断适合度。

这实际上是可行的，但还不足以让我们放弃原来的方法。如果我的论文没有时间限制，我可能已经扩展了算法可用的遗传部分。我很确定我可以大大提高它。

使用的库:如果我没记错的话，用于天文图像数据处理和I/O的IRAF和cfitsio。

在工作中，我遇到了这样一个问题:给定M个任务和N个dsp，如何将任务分配给dsp是最好的?“最佳”定义为“最大负载DSP的负载最小化”。有不同类型的任务，不同的任务类型有不同的性能分支，这取决于它们被分配到哪里，所以我将一组工作到dsp的分配编码为“DNA字符串”，然后使用遗传算法来“培育”我所能“培育”的最佳分配字符串。

它运行得相当好(比我之前的方法好得多，之前的方法是评估每个可能的组合……对于非平凡问题的大小，它将需要数年才能完成!)，唯一的问题是无法判断是否已经达到了最优解。你只能决定当前的“最大努力”是否足够好，或者让它运行更长时间，看看它是否可以做得更好。

我做了一个完整的GA框架，命名为“GALAB”，解决了很多问题:

定位GSM ANTs (BTS)以减少重叠和空白位置。 资源约束项目调度。 进化图景的创造。(Evopic) 旅行推销员问题。 n -皇后和n -颜色问题。 骑士之旅和背包问题。 魔方和数独谜题。 字符串压缩，基于超字符串问题。 二维包装问题。 微型人工生命APP。 鲁比克难题。

足球引爆。我建立了一个GA系统来预测每周澳式足球比赛的结果。

A few years ago I got bored of the standard work football pool, everybody was just going online and taking the picks from some pundit in the press. So, I figured it couldn't be too hard to beat a bunch of broadcast journalism majors, right? My first thought was to take the results from Massey Ratings and then reveal at the end of the season my strategy after winning fame and glory. However, for reasons I've never discovered Massey does not track AFL. The cynic in me believes it is because the outcome of each AFL game has basically become random chance, but my complaints of recent rule changes belong in a different forum.

该系统基本上考虑了进攻强度、防守强度、主场优势、每周的改进(或缺乏)以及这些方面的变化速度。这为每支球队在整个赛季中建立了一组多项式方程。可以计算给定日期的每场比赛的获胜者和分数。我们的目标是找到最接近过去所有游戏结果的系数集，并使用该集合来预测接下来几周的游戏。

在实践中，该系统将找到能够准确预测过去90%以上游戏结果的解决方案。然后，它会成功地为即将到来的一周(即不在训练集中的那一周)挑选大约60-80%的比赛。

结果是:略高于中游水平。没有巨额奖金也没有能打败维加斯的系统。不过很有趣。

我从零开始构建一切，没有使用任何框架。

There was an competition on codechef.com (great site by the way, monthly programming competitions) where one was supposed to solve an unsolveable sudoku (one should come as close as possible with as few wrong collumns/rows/etc as possible).What I would do, was to first generate a perfect sudoku and then override the fields, that have been given. From this pretty good basis on I used genetic programming to improve my solution.I couldn't think of a deterministic approach in this case, because the sudoku was 300x300 and search would've taken too long.