31:T519,md` ### 3. IMITATE: [**Walls and Balls**](https://editor.p5js.org/yangren72/sketches/eWzuZuLxM)

I designed a system that consist of 16 of boxes, each boxes having one moving wall and 100 balls at the begining of each round. The wall's "DNA" have three decisive charactoristic that determine whether it will survive, the moving speed, the length, and the transparency. Each charactoristics of the three has a score, and they are all linked together. The three scored combined is 30, so the higher one charctoristic scores, the lower the other two. The 100 balls will bounce in each box and the walls will try to move in Y-axis bouncing the ball back to prevent the balls to escape from the box. The number of remaining balls will be show in the box. Every 6 seconds, the programe will restart and generate a new generations of eight walls according to how the walls performs in the last round. After a 10 iterations, ideally the walls will evolve to have the DNA that most capible of keeping the balls in the box. The number of boxes, number of balls, duration of each round, and the max iteration can be easily changed on the top section of the program. The results may differ due to the changes of parameters. `32:[{"id":"bd5babec36de934245e63835b05c47f73d1ed27d40c7ba11c7be95c981a6a538959372e696725c7f5e98ee0ba61444e6f17c01d549a64bb693d091eb0c5fffcb","href":"https://static.observableusercontent.com/files/bd5babec36de934245e63835b05c47f73d1ed27d40c7ba11c7be95c981a6a538959372e696725c7f5e98ee0ba61444e6f17c01d549a64bb693d091eb0c5fffcb","name":"es 1.png","type":"image/png","lastModified":1645741521559.888,"size":29299},{"id":"38bf13bf81712c0ebc35c2bba7ecc05f7ec388748117b04f0ad28e646ad06737f9a779066593e5d843341238d6217a746cdc688f50f597c0cf1366b22cec9ef5","href":"https://static.observableusercontent.com/files/38bf13bf81712c0ebc35c2bba7ecc05f7ec388748117b04f0ad28e646ad06737f9a779066593e5d843341238d6217a746cdc688f50f597c0cf1366b22cec9ef5","name":"es2.png","type":"image/png","lastModified":1645741522821.844,"size":418346},{"id":"e35d8afd60a6671069ed0d2a745068b6de843ec9d88439a47a1b3936521200e0d69a0dda77eff4c08ec64550a54ba5d8acfc4cc8424c43dca0b005ab2b968ca2","href":"https://static.observableusercontent.com/files/e35d8afd60a6671069ed0d2a745068b6de843ec9d88439a47a1b3936521200e0d69a0dda77eff4c08ec64550a54ba5d8acfc4cc8424c43dca0b005ab2b968ca2","name":"walls and balls 3.png","type":"image/png","lastModified":1645741521716.791,"size":84855}] 2f:{"title":"Assignment 5: Rule-based systems","cells":[{"id":0,"value":"# Assignment 5: Rule-based systems","pinned":false,"hidden":false,"mode":"md"},{"id":7,"value":"md`\n### 1. READ:\nAfter browsing the articles and conducting some receasrch, I find the **Evolutionary Algorithem** especially inspiring. So I will be trying to imitate it in code.\n

\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":8,"value":"md`\n### 2. DESCRIBE:\nInspired by evolution in nature, **Evolutionary Algorithem** is a system that mutate and elimiate the unfitting genes to evolve a optimal population that can solve combinational problems.\n

\nThere are **four steps** in a EA system.\n

\n**Initialization** is the stage where we create the first generation of population that carries a variety of genes. **Selection** stage is where the population get evaluated with fitness funtion to get a ranking of how well the individuals perform in solving the problem. **Genetic operators** is the stage where we picks the top performed individuals as parents to give birth to the next iterations. It is important to include mutation in the process in order to get the optimal result. Eventually, **Termination** is where the algorithm reached a maximum run time or a certain threshold of performance so it can stop.\n

\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":16,"value":"$31","pinned":false,"hidden":false,"mode":"ojs"},{"id":19,"value":"md`\n### 4. ACTIVATE\n[**Ball Reflecting Walls Evolution System**](https://editor.p5js.org/yangren72/sketches/mtT3iOzZC)\n\n

\n\nI think the evolution algorithm is a simople concept but actually pretty hard to make it work especially in the first time. I really spend a lot of time think about how to make a good system and built it. After finally finish make it work without major issues, I found that it is even harder to make my existing system into an artistic work.\n\nAt the end, I choose to do a more conceptual appraoch. Laying out all the evolved generations on one side, with an a running system on the other side. It is a straight forward display of \"evolutions of ball reflecting walls\". I tweaked the parameter to make the evolution looks more obvious. I also gave the interface a retro look that is inspired by 80s japanese electronics and animations.\n`","pinned":false,"hidden":false,"mode":"ojs"}],"files":"$W32","stdlib":"1","titling":"auto"}