> 2021年08月30日信息消化 ### How To Train Your Brain To Achieve Laser-Like Focus origin: [How To Train Your Brain To Achieve Laser-Like Focus](https://medium.com/mind-cafe/how-to-train-your-brain-to-achieve-laser-like-focus-203a4abee539) #### The Prefrontal Cortex: The Key To Focus The brain has four main regions or lobes: frontal, temporal, parietal, and occipital. The last two perceive our surroundings, whereas the first two — where the prefrontal cortex is located — integrate and analyze the sensory information before making decisions. 大脑有四个主要区域或脑叶：额叶、颞叶、顶叶和枕叶。后两个区域感知我们的周围环境，而前两个区域--前额叶皮层所在的区域--在做决定之前整合和分析感官信息。 In his book, [Memory Rescue: Supercharge Your Brain, Reverse Memory Loss, and Remember What Matters Most](https://www.amazon.com/Memory-Rescue-Supercharge-Reverse-Remember/dp/149642560X), New York Times Bestselling author Dr. Daniel Amen refers to the prefrontal cortex as the brain’s CEO as it “enables us to learn from our mistakes and make plans.” Moreover, it has a central role in our ability to focus. However, the prefrontal cortex can weaken like any muscle, leading to impulsivity, disorganization, poor time management, lack of empathy, among others. For example, though I prided myself in my ability to concentrate through college, after I graduated, I started multitasking, checking social media whenever I wanted, and letting anyone immediate access to my attention. The result? My focus dimmed. At work, I couldn’t stick to a task for longer than fifteen minutes without refreshing my inbox. At home, I could no longer write for hours as I did before. I couldn’t even watch an entire TV episode without checking my phone. ##### How To Train Your Brain’s CEO ###### **Non-work related practices (based on Dr. Amen’s book):** - Language games such as Scrabble and Boggle. - Crossword puzzles. - Speech and debate classes or any other public speaking activities. - Strategy games such as Risk, chess, and Catan. - Prayer and meditation. According to Dr. Amen, “it may be the most powerful prefrontal cortex booster of all. It improves focus, executive function, judgment, and impulse control.” - Weight training combined with aerobic activity (brisk walking). Even better, combine physical activity with mental exercises. - Learning to play a new musical instrument. - Learning memory exercises like those Joshua Foer shares in his book, [*Moonwalking with Einstein: The Art and Science of Remembering Everything.*](https://www.amazon.com/-/es/Joshua-Foer/dp/0143120530/ref=sr_1_1?__mk_es_US=ÅMÅŽÕÑ&dchild=1&keywords=moonwalking+with+einstein&qid=1617961193&s=books&sr=1-1) ###### Work-specific practices (based on Cal Newport’s famous book, Deep Work): - Productive meditation. According to Cal Newport, this type of meditation is when you “take a period in which you’re occupied physically but not mentally — walking, jogging, driving, showering — and focus your attention on **a single well-defined professional problem**. As in mindfulness meditation, you must continue to bring your attention back to the problem when it wanders or stalls.” - [Pomodoro technique](https://en.wikipedia.org/wiki/Pomodoro_Technique#:~:text=The Pomodoro Technique is a,length%2C separated by short breaks.). Though trite, working with intervals of deep work and rest helps us be more productive — and develop future focus. - Environment design. Select a “focus place” you can go to whenever you need to concentrate and rearrange it so that distractions are impossible. For example, I’ve established a rule at home. Whenever the living room — where I write — is closed, my husband can’t enter unless there’s an emergency. Other examples include disabling phone notifications, using noise-canceling headphones, blocking the internet, etc. - Challenging goals. Be specific about what you want to achieve when you sit down to focus. These tasks should be mentally taxing, like writing an article, brainstorming a new product idea, or creating a pitch presentation for potential clients. - Scheduled distractions. Don’t just plan when you’ll do focused work. Arrange times to use social media or other distractions. **Note:** **You can have as many distraction windows as you want. What matters is that you avoid them outside of those periods.** ### 12 Incredible Masterpieces That Show the Oneness of Mathematics and Design origin: [12 Incredible Masterpieces That Show the Oneness of Mathematics and Design](https://medium.com/however-mathematics/12-incredible-masterpieces-that-show-the-oneness-of-mathematics-and-design-7a62ef2f0f51) In his 1969 book, The Science of the Artificial, Herbert Simon defines design as “**To design is to devise courses of action aimed at changing existing situations into preferred ones**.” To move away from the thing at hand to one chosen, the chosen one can’t simply be functional. It has to also appeal to the eye. We can say that things that are functional yet appealing have good design. 赫伯特-西蒙在他1969年出版的《人工科学》一书中，将设计定义为："**设计就是设计行动方案，旨在将现有的情况改变为首选的情况**"。要从手头的东西转移到所选择的东西，所选择的东西不能仅仅是功能性的。它还必须吸引人的眼球。我们可以说，那些既实用又吸引人的东西就是好的设计。 All of the brilliant designers I have had the opportunity to meet had vast imaginations and significant aesthetic concerns. Furthermore, all of them had interests in geometry and mathematics. Spending my best days in Istanbul, the designers at the agency I worked at were having coffee one day. One of them told me, “**When we are designing, we are dealing with mathematics and geometrical shapes**.” That had made me immeasurably happy, as a person with a fantastic imagination and knack for design had stated the opposite of the stereotype of “mathematics is a harsh, mechanical and soulless science” that ordinary people had come up with. 我有机会见到的所有杰出的设计师都有丰富的想象力和重要的美学关注。此外，他们都对几何学和数学有兴趣。在伊斯坦布尔度过我最美好的时光时，我工作的机构的设计师们有一天正在喝咖啡。其中一位告诉我："**我们在设计时，是在与数学和几何图形打交道**"。这句话让我无比高兴，因为一个拥有神奇想象力和设计诀窍的人说出了与普通人得出的 "数学是一门严酷的、机械的、没有灵魂的科学 "的刻板印象恰恰相反。 David Hilbert, one of the best modern mathematicians, thought similarly as well. In David Darling’s, [The Universal Book of Mathematics](https://amzn.to/3ua5Ly5), there is a memory of David Hilbert that is recalled. When he asks the fate of a student who hadn’t attended lectures for a long time, it is said that the a student left his mathematics education to become a poet. In response to this, Hilbert stated, “He made the right decision because he lacked the imagination to be a mathematician.” 大卫-希尔伯特，现代最好的数学家之一，也有类似的想法。在大卫-达林的，[世界数学书](https://amzn.to/3ua5Ly5)中，有一段关于大卫-希尔伯特的记忆被回忆起来。当他问到一个很久没有听课的学生的命运时，有人说，这个学生离开了他的数学教育，成为了一个诗人。对此，希尔伯特说："他的决定是正确的，因为他缺乏成为数学家的想象力"。 Mathematics and design aren’t confined to the golden ratio, however. From the symmetric structure of snowflakes to the fascination of Islamic geometry and the minimalism of Euclidean geometry, many mathematical concepts are present in our lives and nature. All we have to do is read and contemplate. 然而，数学和设计并不局限于黄金比例。从雪花的对称结构到伊斯兰几何的魅力和欧几里得几何的极简主义，许多数学概念都存在于我们的生活和自然中。我们所要做的就是阅读和沉思。 ##### [**Poemotion Series by Takahiro Kurashima**](https://amzn.to/2QMHz6f)  Kurashima has prepared incredibly interactive book with the inspiration he gathered from Moiré patterns. The designs, which have a semi-transparent foil over them, give you a feeling of a magical moment when you are surprised by the moving geometric shapes they change into. You can momentarily feel your brain burning. ``` Moire patterns, by definition, are the patterns that show up when you overlay, then slightly move around two periodic designs. For example, Moire patterns can be found in computer monitors. When you mess around with the video settings, the resulting blurriness is a result of Moire patterns. ``` > In mathematics, physics, and art, moiré patterns (UK: /ˈmwɑːreɪ/ MWAR-ay, US: /mwɑːˈreɪ/ mwar-AY,[1] French: [mwaʁe] (About this soundlisten)) or moiré fringes[2] are large-scale interference patterns that can be produced when an opaque ruled pattern with transparent gaps is overlaid on another similar pattern. For the moiré interference pattern to appear, the two patterns must not be completely identical, but rather displaced, rotated, or have slightly different pitch. > > 在数学、物理学和艺术中，摩尔纹图案（英国。/ˈmwɑːreɪ/ MWAR-ay, US: /mwɑːˈreɪ/ mwar-AY, [1] French: [mwaʁe]（关于这个声音列表））或摩尔纹[2]是大规模的干扰模式，当一个不透明的统治模式与透明的间隙被覆盖在另一个类似的模式可以产生。为了使莫尔雷干涉图案出现，这两个图案必须不是完全相同的，而是移位、旋转或有稍微不同的间距。 ##### [**Oliver Byrne: The First Six Books of the Elements of Euclid by TASCHEN**](https://amzn.to/3eOxQUW) Euclid is the father of geometry. He is the first to formulate the sentence, “Only one line can be drawn between two points.” Furthermore, as in his own words, Euclid didn’t just want to believe in truths; he strived to prove and know them. His obsession with the absolute truth is what made him the father of geometry. 欧几里德是几何学之父。他是第一个提出 "两点之间只能画一条线 "这句话的人。此外，正如他自己所说的那样，欧几里德不只是想相信真理；他还努力证明和认识这些真理。他对绝对真理的痴迷，使他成为几何学之父。 Euclid wanted all of humanity to learn of the findings he had discovered with only a compass and a ruler. That is why he wrote a series of books, stretching six volumes, called ‘**Elements**.’ The books he wrote thousands of years ago still make up the foundation of geometry taught in schools today. In his article, ‘On The Method of Theoretical Physics,’ Einstein states, “**If Euclid failed to kindle your youthful enthusiasm, then you were not born to be a scientific thinker.**” One of the greatest thinkers of modern times, Bertrand Russell, identifies his discovery of ‘Elements’ as “one of the greatest events in my life, as dazzling of first love.” 欧几里德希望全人类都能了解他只用一个罗盘和一把尺子就能发现的结果。这就是为什么他写了一系列的书，绵延六卷，称为'**元素'。他在几千年前写的书仍然构成了今天学校教授的几何学的基础。爱因斯坦在他的文章《论理论物理学的方法》中指出："如果欧几里得未能点燃你年轻时的热情，那么你就不是天生的科学思想家"。现代最伟大的思想家之一伯特兰-罗素将他对'元素'的发现认定为 "我生命中最伟大的事件之一，就像初恋一样耀眼"。  Many years after Euclid, in 1847, an Irish geometry teacher, Oliver Byrne, released Euclid’s Elements in a beautiful red, blue, and yellow form. To this day, Oliver Byrne’s book is thought to be the most attractive edition of Euclid’s Elements by many. 在欧几里德之后许多年，1847年，一位爱尔兰几何学教师奥利弗-伯恩（Oliver Byrne）以美丽的红、蓝、黄三色发布了《欧几里德原理》。时至今日，奥利弗-伯恩的书被许多人认为是欧几里德《元素》最吸引人的版本。 ##### [**Overview: A New Perspective of Earth by Benjamin Grant**](https://amzn.to/3nI7vvV)  ‘Overview’ is a book that will entirely change the way you view our planet. Although the satellite images that Benjamin Grant has gathered from Google Earth may seem simple at first, each of them is a geometric pattern in and of itself. This book proves that our planet is truly a work of geometric art. ##### [**The ABC’s of Triangle, Square, Circle by Ellen Lupton**](https://amzn.to/2QPMaVg) The Bauhaus was one of the best design and architecture schools in Germany in the 1900s. This book is about the lessons taught at The Bauhaus.The book explains the German design era of 100 years ago and the ideas that lead to it with incredible images and detailed graphics. The ABC’s of Triangle, Square, Circle is a piece that should be used for design and the teaching of geometry. That is because some of the passages in book are eye-opening. ##### [**The Golden Ratio: The Divine Beauty of Mathematics**](https://amzn.to/3te2WL0)  This book proves that what makes nature, architecture, and art pretty is what the human brain will always find beautiful, the golden ratio. It does this in a straightforward tongue as well. That means even those who are not very well versed mathematically can understand the technical parts of the book. It can even be used as a great conversation starter in your friend groups. ### Documentation Guide for Developers origin: [Documentation Guide for Developers](https://hellonehha.hashnode.dev/documentation-guide-for-developers-cksie1jh7050pvps1bnrgf7vh?source=newsletter) > sDocumentation required 10-20% time effort, but the impact or use of it is big or immense #### ???? What to document? ##### 1. Feature and approaches You can start documenting the approaches, LLD, HLD of the solution of the feature. You can add: - different solutions - prefer a solution with reasons - detailed design of the solution - dependencies - APIs contracts - any trade-offs - LLD (Low-Level Design) - HLD (High-Level Design) ???? [developer-story-template.md](https://github.com/Neha/documentation/blob/master/developer-story-template.md) ##### 2. Knowledge sharing This is one of the most underrated at work. Document the knowledge sharing. It could have: - Early review of any tech - Anything you learned - A bug you fixed - A feature we should have - A complex feature ???? [knowledge-sharing-template.md](https://github.com/Neha/documentation/blob/master/knowledge-sharing-template.md) ##### 3. Code guidelines If you are a lead, Senior developer, or Engineering Manager then this is 1st documentation you should have. Code guidelines are helpful for the new devs joining the team, set a bar, and most important reduce the code-review comments (logical comments will stay). After a few revisions, you will see the value in this documentation. We can automate the code guidelines in the project by using CI/CD, linters, and npm packages will help you in automating quite a few things. ???? [code-guidelines-template.md](https://github.com/Neha/documentation/blob/master/code-guidelines-template.md) ##### 4. Checklist One constant thing in every project (small/big) is missing something while doing the code deployment. I would say it is a must to have a checklist of the things to take care of before doing deployment or going Live. ???? [checklist-template.md](https://github.com/Neha/documentation/blob/master/checklist-template.md) ##### 5. API Contracts One of the common ways of working is in collaboration with the APIs. Though it is expected that the API team must be documenting their contracts if they are not then you should suggest them to do. Eg: [Stripe](https://stripe.com/docs/api), [PayPal](https://developer.paypal.com/docs/api/overview/) This should happen from Day-0. Again, the outline would be just like a developer conversation: - Problem - Suggestions - Solutions pros & cons - trade-offs - challenges - feedback ???? [Sample Template](https://github.com/Neha/documentation/blob/master/api_template.md) ##### 6. Dependencies While working on the big/large-scale projects or any size of the project it is common to have dependencies on the different teams, code such as APIs, attribute(s), backend, DevOps, etc. I would say it is MUST document the dependencies of your code-base and project. Your future self and your team will be thankful to you ???? ???? [dependencies_template.md](https://github.com/Neha/documentation/blob/master/dependencies_template.md) > As an Engineering Manager, allow developers an opportunity to try the different features to know the code-base. The most common problem I have seen is that the new developer is not aware of the code's dependencies. Documentation such as this would be super helpful. Also, This is also an opportunity for developers to push their code to quality by reviewing the dependencies, and if there is something the developer can fix something. ##### 7. Sprint Milestones, celebrations, retrospect, and reviews As a senior developer, tech lead, or Engineering manager block time in your team and at the end of your sprint present sprint milestones, review, and celebrate. Yes, we do have JIRA, yes, we do have the scrum but as a developer doing a playback (playback of the work done in the last sprint) is important. Why? it is an opportunity to appreciate the good work, celebrate the milestones, and reflect back on mistakes and retrospect. This document could be a reference when a team wants to see how far we have come, what we have delivered, and milestones achieved. ???? [sprint-review.md](https://github.com/Neha/documentation/blob/master/sprint-review.md) ##### 8. Cookbooks A cookbook is a beautiful way to set the standard guidelines across organizations from the engineering side. For example, a cookbook could have the performance guide, metrics, measurement of front-end apps, back-end, API. A cookbook is a 'black book' for your developers to see the area to focus on, what is expected, how to achieve, etc. As a tech lead or engineering manager , investing time in the Cook Book is a good investment. The cookbook helps the developers what is expected from their contribution to the product/project, how the review of their work will be done or how would be measure. Eg: Accessibility should be at least AA level. If a developer's code is meeting that then it is a concern. Similarly, these cookbooks would help in quantifying the quality of the code. ##### 9. Brag Document A brag document is a personal document that every developer should maintain. This document will have all the things you have achieved - small/big. Basically, everything which created an impact - code, tech discussion, mentoring, hiring, tech talk, etc. At the time of your yearly review, you can refer to this document to look back at your progress, share with your leads, or whenever you feel low you can just go through this doc to reflect back. ##### 9. Project Starter Guide Create a project starter guide for your project. This document will reduce the repetitive work/meetings/communication. This document could covers: - Summary of your project - Team structure - Product/Project walkthrough - WOW (way of working) - Code walkthrough - Tools and Applications required - Escalation Process - and anything which is required in your project - HLD (High-Level Design) ##### ???? Where to document? - Wiki (GitHub) - confluences (Atlassian) - Drive (any Google, One Drive) - or a shared folder with readme(s), or doc(s) ### This 2,500 Year Old Technique is the Secret Behind Super Human Memory origin: [This 2,500 Year Old Technique is the Secret Behind Super Human Memory](https://daveasprey.com/memory-palaces-improve-memory/?utm_source=ActiveCampaign&utm_medium=email&utm_content=This+2%2C500+Year+Old+Technique+is+the+Secret+Behind+Super+Human+Memory&utm_campaign=This+2%2C500+Year+Old+Technique+is+the+Secret+Behind+Super+Human+Memory&vgo_ee=01HGzkI7yMvmAnZP6gYDPp%2FvRHj2NQDVX9U4hkSVncU%3D) Imagine, for a moment, that there were a simple technique out there that could dramatically improve your memory. I’m not talking about an improvement of 100%, 200%, or even 300%. I’m talking about allowing you to remember anywhere between 10-100x more information. Though you may have heard of The Memory Palace (or “the method of loci”), most people have never created one – much less turned it into a habit. And yet, memory palaces are, without a doubt, the single most powerful thing you can do to enhance your memory. (In [my latest book](https://amzn.to/37h8Q37), I refer to them as “the mnemonic nuclear option,” a bit like dropping a nuke on a schoolyard bully). Researchers have actually proven that this technique reshapes neural networks to support superior memory.[[1\]](https://daveasprey.com/memory-palaces-improve-memory/?utm_source=ActiveCampaign&utm_medium=email&utm_content=This+2%2C500+Year+Old+Technique+is+the+Secret+Behind+Super+Human+Memory&utm_campaign=This+2%2C500+Year+Old+Technique+is+the+Secret+Behind+Super+Human+Memory&vgo_ee=01HGzkI7yMvmAnZP6gYDPp%2FvRHj2NQDVX9U4hkSVncU%3D#ref-list) 虽然你可能听说过 "记忆宫殿"（或 "oci方法"），但大多数人从未创建过一个，更不用说把它变成一种习惯。然而，毫无疑问，记忆宫殿是你能做的提高记忆力的唯一最有力的事情。(在[我的新书](https://amzn.to/37h8Q37)中，我把它们称为 "记忆的核选择"，有点像对校园恶霸投掷核弹）。) 研究人员实际上已经证明，这种技术可以重塑神经网络以支持卓越的记忆。[[1]](https://daveasprey.com/memory-palaces-improve-memory/? utm_source=ActiveCampaign&utm_medium=email&utm_content=This+2%2C500+Year+Old+Technique+is+the+Secret+Behind+Super+Human+Memory&utm_campaign=This+2%2C500+Year+Old+Technique+is+the+Secret+Behind+Super+Human+Memory&vgo_ee=01HGzkI7yMvmAnZP6gYDPp%2FvRHj2NQDVX9U4hkSVncU%3D#ref- 列表) They’ve even discovered that these changes are long-lasting… and that anyone can use them to transform their brain into that of a memory champion.[2] In other words, unlike the Olympic games, the only differences between you and Memory Games champions are technique and practice. 他们甚至发现，这些变化是持久的......而且任何人都可以利用它们将自己的大脑转变为记忆冠军的大脑。[2] 换句话说，与奥林匹克运动会不同，你和记忆运动会冠军之间唯一的区别是技术和实践。 ##### How (And Why) The Memory Palace Actually Works The idea behind The Memory Palace technique is very simple. First, choose a location that is familiar to you. This can be a past or current home, an office, a friend’s place, or even a store you frequent. I suggest choosing the location based on how much information you wish to remember. After all, you wouldn’t want to waste a 5-story office building to memorize the 45 U.S. presidents or the NATO Phonetic alphabet. 记忆宫殿技术的理念非常简单。首先，选择一个你熟悉的地点。这可以是过去或现在的家，办公室，朋友的地方，甚至是你经常去的商店。我建议根据你希望记住的信息的多少来选择地点。毕竟，你不会想浪费一栋5层楼的办公楼来记忆45位美国总统或北约的拼音字母。 Once you have your location picked out, decide on a “path” that you are going to take as you walk through it. I recommend starting at the entrance and walking along the “perimeter” of the building, tracing along the walls either clockwise or counter-clockwise. 一旦你选好了地点，就决定你要走的一条 "路"，因为你要穿过它。我建议从入口处开始，沿着建筑物的 "周边 "走，沿着墙壁顺时针或逆时针追踪。 Trace the path that you’d use to go into each room, and make sure that you never cross your own path. This is less important if you’re going to be memorizing non-sequential information, such as foreign language vocabulary. It’s absolutely essential for things like speeches, decks of cards, and so on. 追踪你进入每个房间的路径，并确保你不会与自己的路径交叉。如果你要记忆非连续的信息，如外语词汇，这就不那么重要了。对于像演讲稿、扑克牌之类的东西，这是绝对必要的。 If your information is non-sequential, decide how you’ll break it up into groups, and which rooms in the palace will correspond to those groups. (You can do this in your head, or by sketching out a floor plan on a piece of paper, if it’s easier). 如果你的信息是不连续的，决定你如何将其分成几组，以及宫殿中的哪些房间将与这些组相对应。(你可以在脑子里做这件事，或者在一张纸上画出一个平面图，如果这样做更容易的话）。) With your memory palace set up, it’s time to get visual. Create a novel, bizarre, and creative mental image for the first piece of information you want to remember. In the SuperLearner® methodology, we call these “markers,” and they’re the building block of all the mnemonic techniques we use. To make your markers extra-memorable, you’ll want to combine elements of your existing memories in ways that are particularly strange – even violent or sexual. 在你的记忆宫殿建立起来后，是时候进行视觉化了。为你想记住的第一条信息创造一个新奇、怪异、有创意的心理形象。在SuperLearner®方法论中，我们称这些为 "标记"，它们是我们使用的所有记忆技术的基础。为了使你的标记特别容易记住，你要把你现有记忆中的元素以特别奇怪的方式--甚至是暴力或性的方式结合起来。 Imagine, for example, I wanted to learn the Spanish word for “stove” (estufa). I might visualize actress Gloria Estefan sitting on my kitchen stove and shrieking in pain as her backend sizzles. This image is great because it combines my own existing knowledge of Gloria Estefan with the location of the stove in my home – in a way that will be hard for most people to forget. Once I can “see” that particularly bizarre image in my mind’s eye, my work is done. I can now move on to my next location, whether that’s the corner of a bed, a painting on the wall, a window sill, or a bookshelf, and place another marker there. 例如，想象一下，我想学习西班牙语中的 "炉子"（estufa）这个词。我可能会想象女演员格洛丽亚-埃斯特凡（Gloria Estefan）坐在我的厨房炉子上，当她的后背被烤焦时痛苦地叫喊着。这个图像很好，因为它将我自己对格洛丽亚-埃斯特凡的现有知识与我家炉子的位置结合起来--以一种大多数人很难忘记的方式。一旦我能够在我的脑海中 "看到 "那个特别奇怪的图像，我的工作就完成了。我现在可以转到下一个地点，不管是床角、墙上的画、窗台，还是书架，在那里再放一个标记。 Now I know what you’re thinking: “how do I come up with these visual markers?” Of course, there are different techniques for different types of information. For strings of numbers, you would use a system like The Major Method[4] – or a much more complex system if you were looking to compete. For things like names, speeches, bible verses, and so on, you can get creative. “Mary” becomes a visualization of the Virgin Mary. Ray becomes a Manta Ray. You get the idea. Markers are highly individual, because the best ones integrate your own pre-existing memories. But don’t worry, the actual contents of your markers don’t matter so much as the fact that you create them. 现在我知道你在想什么。"我是如何想出这些视觉标记的？" 当然，对于不同类型的信息有不同的技巧。对于一串数字，你会使用像 "主要方法"[4]这样的系统--如果你想竞争的话，也可以使用更复杂的系统。对于像人名、演讲稿、圣经经文等等的东西，你可以发挥创意。"玛丽 "成为圣母玛利亚的形象化。雷变成了曼塔雷。你会明白的。标记是高度个性化的，因为最好的标记整合了你自己预先存在的记忆。但不要担心，你的标记的实际内容并不重要，重要的是你创造了它们。 Once you’ve stocked your memory palace full of visualizations, you’re done! All you need to do is periodically drop in for a visit and review it to prevent your brain from forgetting it. Realistically, though, it doesn’t take much. I’ve had people approach me at events and tell me about that “annoyingly sticky” memory palace I had them memorize 5 years ago – and how it still lingers on in their mind. 一旦你在你的记忆宫殿里放满了视觉化的东西，你就完成了！你需要做的就是定期对你的记忆宫殿进行检查。你所需要做的就是定期去看一看，回顾一下，防止你的大脑忘记它。不过，从现实的角度来看，这并不需要太多。我曾在活动中遇到过一些人，告诉我五年前我让他们记忆的那个 "令人讨厌的粘性 "记忆宫殿--以及它如何在他们的脑海中挥之不去。 ##### Now You Try It As any coach worth their salt will tell you, the information above is not enough to create a transformation. After all, you can’t learn how to swim in a library. So if you actually want to reap the benefits of the Memory Palace, you’re going to have to try it out for yourself. Perhaps you already know exactly what you’d like to memorize. Hey, that’s great! Go do it, and [let me know how it goes](https://twitter.com/entreprenewer). But for those of you who are scratching your heads on where to begin, here are some of my favorite “beginner” memory palaces that always come in handy: - [The NATO Phonetic Alphabet](https://en.wikipedia.org/wiki/NATO_phonetic_alphabet) - The [Circle of Fifths (Music Theory)](https://www.youtube.com/watch?v=mrvCCWY9D74) - Your grocery list - The names, faces, and locations of everyone at your company or division - The first 25 digits of Pi - Your credit card numbers #### An Introduction to AI Story Generation origin: [An Introduction to AI Story Generation](https://thegradient.pub/an-introduction-to-ai-story-generation/) In addition to these fundamental AI research problems, automated story generation is also worth studying for the applications it may enable. Aside from the grand challenge of an AI system that can write a book that people would want to read, storytelling appears in many places in society: - **Human-AI coordination**: there are times when it is easier to communicate via narrative. For example, communicating via vignettes helps with coordination because it sets expectations against which to gauge the appropriateness of behavior. Humans often find it easier to explain via vignettes, and are often able to more easily process complex procedural information via vignettes. - **人与人工智能的协调**：有些时候，通过叙事来沟通更容易。例如，通过小故事进行沟通有助于协调，因为它设定了期望值，据此来衡量行为的适当性。人类经常发现通过小故事进行解释更容易，而且往往能够通过小故事更容易地处理复杂的程序性信息。 - **Human-AI rapport:** Telling and listening to stories is also a way that humans build rapport. **人与人工智能的关系：**讲故事和听故事也是人类建立关系的一种方式。 - **Explainable AI:** Explanations can help humans understand what an AI system does. For sequential decision making tasks (e.g. robotics) this might entail a temporal component to the explanation resembling a story. **可解释的人工智能：**解释可以帮助人类理解人工智能系统的工作。对于连续的决策任务（如机器人），这可能需要在解释中加入类似于故事的时间成分。 - **Computer games**: many computer games feature stories or plots, which can be generated or customized. Going beyond linear plots, interactive stories are those in which the user assumes the role of a character in a story and is able to change the story with their actions. To be able to respond to novel user actions requires the ability to adapt or re-write the plot. **电脑游戏**：许多电脑游戏都有故事或情节，可以生成或定制。超越了线性情节，交互式故事是指用户在故事中扮演一个角色，并能够通过他们的行动来改变故事。为了能够对新奇的用户行动作出反应，需要有能力调整或重写情节。 - **Training and education:** inquiry-based learning puts learners in the role of experts and scenarios can be generated to meet pedagogical needs (similar to interactive stories above). **培训和教育：**基于探究的学习使学习者处于专家的角色，并且可以生成场景以满足教学需要（与上述互动故事类似）。 ##### Non-Learning Story Generation Approaches Let’s get into technologies. This cannot be exhaustive, so I have attempted to create some broad classes and give some examples of each. This section looks at non-machine-learning based approaches. Non-learning systems dominated much of the history of automated story generation. They can produce good plots, though the emphasis on natural language output has been reduced. The key defining feature of these techniques — for the most part — is the reliance on knowledge bases containing hand-coded knowledge structures. ###### Story Grammars Computational grammars were designed to decide whether an input sequence would be accepted by a machine. Grammars can be reversed to make generative systems. The earliest known story generator (Grimes 1960) used a hand-crafted grammar. The details are largely lost to history. In 1975,[ David Rumelhart (1975)](https://www.sciencedirect.com/science/article/pii/B9780121085506500136) published a grammar for story understanding. It was followed by a proposed story grammar by[ Thorndyke (1977](https://www.sciencedirect.com/science/article/abs/pii/0010028577900056)).  Structural narratology, drawing heavily from[ Bal (1998)](https://books.google.com/books/about/Narratology.html?id=jPj4Bq0H4JoC), analyze narratives at two levels: - **Fabula:** The fabula of a narrative is an enumeration of all the events that occur in the story world between the time the story begins and the time the story ends. The events in the fabula are temporally sequenced in the order that they occur, which is not necessarily the same order in which they are told. Most notably, the events in the fabula might not all exist in the final telling of the narrative; some events might need to be inferred from what is actually told. For example: “John departs his house. Three hours later John arrives at the White House. John mutters about the traffic jam.” The fabula clearly contains the events “John departs house” and “John arrives at the White House” and “John mutters”. We might infer that John also drove a car and was stuck in a traffic jam — an event that was not explicitly mentioned and furthermore would have happened between “depart” and “arrive” instead of afterward when the first clue is given. 故事集。叙事诗是对故事开始到故事结束期间发生在故事世界里的所有事件的列举。故事中的事件在时间上是按其发生的顺序排列的，这不一定与故事的顺序相同。最值得注意的是，故事中的事件不一定都存在于最后的叙述中；有些事件可能需要从实际讲述的内容中推断出来。比如说。"约翰离开了他的房子。三小时后，约翰到达白宫。约翰嘀咕着交通堵塞的问题"。这个故事显然包含了 "约翰离开家 "和 "约翰到达白宫 "以及 "约翰喃喃自语 "这些事件。我们可以推断出，约翰也开着车，被堵在了路上--这个事件没有被明确提及，而且会发生在 "出发 "和 "到达 "之间，而不是在给出第一条线索之后。 - **Sjuzhet:** The sjuzet of a narrative is a subset of the fabula that is presented via narration to the audience. It is not required to be told in chronological order, allowing for achronological tellings such as flash forward, flashback, ellipses (gaps in time), interleaving, achrony (randomization), etc. 叙事词（Sjuzhet）。叙述的sjuzet是fabula的一个子集，通过叙述呈现给观众。它不需要按照时间顺序来讲述，允许采用倒叙、闪回、省略号（时间上的空白）、交错、随机化（随机化）等非顺叙方式。 ##### Controllable Neural Story Generation One of the main limitations of neural language models is that they generate tokens based on a sequence of previous tokens. Since they are backward-looking instead of forward-looking, there is no guarantee that the neural network will generate a text that is coherent or drives to a particular point or goal. Furthermore, as the story gets longer, the more of the earlier context is forgotten (either because it falls outside of a window of allowable history or because neural attention mechanisms prefer recency). This makes neural language model based story generation systems “fancy babblers” — the stories tend to have a stream-of-consciousness feel to them. Large-scale pre-trained transformers such as GPT-2, GPT-3, BART, and others have helped with some of the “fancy babbling” issues by allowing for larger context windows, but the problem is not completely resolved. As language models themselves they cannot address the problem of forward-looking to ensure they are building toward something in the future, except by accident. 神经语言模型的主要局限性之一是，它们是根据以前的标记序列来生成标记的。由于它们是向后看的，而不是向前看的，所以不能保证神经网络会产生一个连贯的文本，或驱动到一个特定的点或目标。此外，随着故事越来越长，早期的背景被遗忘得越多（要么是因为它不在可允许的历史窗口内，要么是因为神经注意机制更倾向于回顾性）。这使得基于神经语言模型的故事生成系统成为 "花哨的胡言乱语者"--这些故事往往有一种意识流的感觉。大规模的预训练转化器，如GPT-2、GPT-3、BART等，通过允许更大的上下文窗口，帮助解决了一些 "花式胡言乱语 "的问题，但这个问题并没有完全解决。作为语言模型本身，它们不能解决前瞻性的问题，以确保它们是朝着未来的东西建立的，除非是意外。 Tambwekar et al. (2019) use reinforcement learning to fine-tune a sequence-to-sequence language model to generate story continuations that move toward a given goal. Reinforcement learning, generally, is a technique that can be used to solve sequential decision-making problems. However the latent space of a language model is too large for true trial-and-error learning. The reinforcement learner is used as a non-differentiable loss function. But how does the reinforcement learner know whether the continuation is getting closer to a particular goal? The system extracts patterns of verbs from the story corpus, clusters them according to how far each verb is from the goal, and then rewards the language model when it produces a continuation with a verb in the next cluster closer to the goal. Tambwekar等人（2019年）使用强化学习来微调一个序列到序列的语言模型，以生成朝向特定目标的故事延续。强化学习，一般来说，是一种可以用来解决顺序决策问题的技术。然而语言模型的潜在空间对于真正的试错学习来说太大。强化学习器是作为一个无差别的损失函数来使用的。但是，强化学习器如何知道延续性是否正在接近一个特定的目标？该系统从故事语料库中提取动词的模式，根据每个动词离目标的远近对它们进行聚类，然后当语言模型产生的续篇中有一个动词在下一个聚类中更接近目标时，就对它进行奖励。 While typically language models are given a prompt that is the first line of the story, neural language models can also be conditioned on plot points that need to be present in the story. This increases the probability that certain events will occur as the story generation progresses. The[ **hierarchical fusion model**](https://arxiv.org/abs/1805.04833) ([Fan et al. 2018](https://arxiv.org/abs/1805.04833)) takes a one-sentence description of the story content and produces a paragraph. [**PlotMachines**](https://arxiv.org/abs/2004.14967) ([Rashkin et al. 2020](https://arxiv.org/abs/2004.14967)) conditions a generator on a set of concept phrases given by a user. However, instead of using this set of concept phrases as an outline to elaborate on, the system can decide for itself what order to introduce the concepts. Instead of being given a high-level plot, a system can first try to generate a high-level plot and then elaborate. The[ **plan-and-write**](https://arxiv.org/abs/1811.05701) technique ([Yao et al. 2019](https://arxiv.org/abs/1811.05701)) also uses a two-level approach. Instead of a single high-level sentence, the plan-and-write system learns to generate a sequence of keywords. Each of these keywords is then used to condition a language model so that it produces content about that keyword. If the keywords present a coherent progression then it stands that the lower-level elaborated content will as well. The technique by [Fan, Lewis, and Dauphin (2019)](https://arxiv.org/abs/1902.01109) has one model trained on entities and relations extracted from stories and a second natural language generator conditioned on the outputs of the first model. 虽然通常语言模型得到的提示是故事的第一行，但神经语言模型也可以以故事中需要出现的情节点为条件。这增加了某些事件在故事生成过程中发生的概率。分层融合模型（Fan等人，2018）采取对故事内容的一句话描述，并产生一个段落。PlotMachines（Rashkin等人，2020）将生成器的条件放在用户给出的一组概念短语上。然而，系统不是把这组概念短语作为大纲来阐述，而是可以自己决定以什么顺序来介绍这些概念。系统可以首先尝试生成一个高层次的情节，然后再进行阐述，而不是被给定一个高层次的情节。计划与写作技术（Yao等人，2019年）也采用了两级方法。计划和写作系统不是一个单一的高级句子，而是学习生成一连串的关键词。然后，这些关键词中的每一个都被用来调节语言模型，使其产生关于该关键词的内容。如果关键词呈现出连贯的进展，那么较低层次的阐述内容也会如此。Fan、Lewis和Dauphin（2019）的技术有一个根据从故事中提取的实体和关系训练的模型，以及第二个以第一个模型的输出为条件的自然语言发生器。  ##### Neuro-Symbolic Generation One of the issues with neural language models is that the hidden state of the neural network (whether a recurrent neural network or a transformer) only represents what is needed to make likely word choices based on a prior context history of word tokens. The “state” of the neural network is unlikely to be the same as the mental model that a reader is constructing about the world, focusing on characters, objects, places, goals, and causes. The shift from symbolic systems to neural language models shifted the focus from the modeling of the reader to the modeling of the corpus. This makes sense because data in the form of story corpora is readily available but data in the form of the mental models readers form is not readily available. 神经语言模型的一个问题是，神经网络（无论是递归神经网络还是转化器）的隐藏状态只代表了根据先前的词汇标记的上下文历史做出可能的词汇选择所需要的东西。神经网络的 "状态 "不太可能与读者正在构建的关于世界的心理模型相同，它侧重于人物、物体、地点、目标和原因。从符号系统到神经语言模型的转变，将重点从读者的建模转移到语料库的建模。这是有道理的，因为故事语料库形式的数据是现成的，但读者形成的心理模型形式的数据却不是现成的。 Assuming the theories about how reader mental models can be represented symbolically are correct, can we build neurosymbolic systems that take the advantages of neural language models and combine them with the advantages of symbolic models? Neural language models gave us a certain robustness to a very large space of inputs and outputs by operating in language instead of limited symbols spaces. But neural language model based story generation also resulted in a step backward from the perspective of story coherence. Symbolic systems on the other hand excelled at coherence through logical and graphical constraints but at the expense of limited symbol spaces. 假设关于读者心智模型如何被符号化的理论是正确的，我们能否建立神经符号系统，将神经语言模型的优势与符号模型的优势结合起来？神经语言模型通过在语言而不是有限的符号空间中操作，给我们提供了对非常大的输入和输出空间的某种稳健性。但是基于神经语言模型的故事生成也导致了从故事连贯性的角度来看的退步。另一方面，符号系统通过逻辑和图形约束在一致性方面表现出色，但却牺牲了有限的符号空间。 Lara Martin (Dissertation, 2021) proposes to take a neural language model such as GPT-2 and constrain it with reasoning about causality. GPT-2, as a language model, can generate story continuation probabilistically based on a history of prior word tokens. Martin’s system parses the generated continuation sentence and uses VerbNet to infer what would be known to readers about the preconditions and effects of the sentence. If the preconditions of the continuation are not supported by the effects of prior sentences in the story, the continuation is rejected. If the preconditions are supported, then the effects of the sentence are used to update a set of logic-like prepositions that describe the world. These prepositions, which are drawn from VerbNet exactly because they are based on what readers can infer from reading a sentence, can be thought of as a simple reader model. By tracking the hypothetical reader’s model this system is less likely to make transitions from one event to another that do not make sense to the reader. 拉拉-马丁（论文，2021年）提议采用GPT-2这样的神经语言模型，并用因果关系的推理来约束它。GPT-2作为一个语言模型，可以根据先前的单词标记的历史，以概率的方式生成故事的延续性。马丁的系统解析生成的续句，并使用VerbNet来推断读者对该句子的前提条件和影响会有什么认识。如果续句的前提条件没有得到故事中先前句子效果的支持，续句就被拒绝。如果前提条件得到支持，那么该句子的效果就会被用来更新一套描述世界的类似逻辑的介词。这些介词正是从VerbNet中提取的，因为它们是基于读者从阅读一个句子中可以推断出的内容，可以被认为是一个简单的读者模型。通过跟踪假设的读者模型，这个系统不太可能从一个事件过渡到另一个事件，而这对读者来说是没有意义的。