AI - York Jong on Diigo

23 Jul 09

Antiobjects - Wikipedia, the free encyclopedia

The notion of antiobjects is a computational metaphor useful to conceptualize and solve hard problems by swapping computational foreground and background.

en.wikipedia.org/Antiobjects - Preview

algorithm maze game AI

The notion of antiobjects is a computational metaphor useful to conceptualize and solve hard problems by swapping computational foreground and background.
Antiobjects turn things on their head. In the case of Pacman we put the main computation into the maze; to simulate the behavior of an air bubble we put the main computation into the water; to create a collaborative soccer game we put the main computation into the soccer field.
2 more annotations...
- These local differences affect the outcome of each tile's amount of scent and always provide a gradient that leads to Pac-Man.
- The Ghosts are given a simple hill climbing algorithm to walk towards higher quantities of Pac-Man scent.

20 Jun 09

Introduction to AI Robotics

The book is a broad survey and represents the state of the practice circa 1999. Robotics is a rapidly changing field and I am working on the second edition this year. I teach the course each year in the fall to approximately 35 undergraduate and 10 graduate students and update the materials accordingly. In the meantime, my slides reflect advances and alternative organization of themes. Probably the biggest changes are to the sections on teleoperation, multi-agents, and human-robot interaction, and the de-emphasis on architectures.

www.csee.usf.edu/book - Preview

book choice robot embedded AI

22 May 09

Flesh and Machines 我們都是機器人

cilab.csie.ncu.edu.tw/...flesh_and_machi_1.html - Preview

3star AFSM AI behavior-based book choice concept iRobot robot subsumption

21 May 09

科幻中的非人類

在科幻電影或小說裏，都經常出現一些「非人類」的物種，由我們親手創造到外星人都有。而在人類與這些非人類之間的矛盾，通常都會引發衝突和戰爭，正如現實世界一樣。彼此信念不同，也不肯讓步，其所引起的紛爭，就成為科幻片的其中一個恒常的主題。而這些非人類，又大概可以分為以下類別：

jacky.seezone.net/android - Preview

sci-fi robot AI

18 May 09

智慧型控制：分析與設計

「智慧型控制」成為當下日常生活切身話題，舉凡家電、醫療、國防等隨處可見。本書可分三大部份，第一部神經網路控制，第二部模糊邏輯控制，第三部基因演算控制。書中對於上述三種控制法理論、方法及應用有詳細介紹。對於相關的數學基礎，作者以最淺顯易懂方式來論述，希望讀者在閱讀此書時能以最快、最清楚方式進入智慧型控制的殿堂，是一本適合大學、科大電機系「智慧型控制」課程。

findbook.tw/...basic - Preview

book AI toread

06 May 09

創智慧-理解人腦運作，打造智慧機器

智慧究竟是什麼？二十多年來人工智慧和神經網路等想要模擬人類智慧的嘗試失敗，正是因為他們沒有先了解什麼是智慧。霍金斯提出一個強有力的理論，解釋人類大腦是怎樣運作以及為什麼他認為電腦沒有智慧，深入探討如何建構出真正的智慧機器。

www.books.com.tw/...booksfile.php - Preview

book choice brain AI

群智能算法及其应用

www.haoplus.com/...1394620.html - Preview

AI Swarm ant book choice particle complexit

17 Feb 08

《机器人》第三章－能力和显现

www.yrobot.net/...71_820.html - Preview

5star AI book computer memory robot trend

三十年的计算机视觉的历程表明，1 MIPS的机器能从实时图像中提取简单的特征量───在杂色背景里跟踪白线或白点。10 MIPS的机器能跟随复杂的灰度区，巡航导弹、灵巧炸弹和早期的自驾驶大蓬车证实了这点。100 MIPS的机器恰好能追踪路面上不可预测的特征量，Navlab在长距离的行程证实了这点。1000 MIPS的机器完全能识别三维空间中带纹理的粗糙物体，某些中分辨率的立体视觉程序解释了这个问题，也包括我的程序。几个“垃圾箱挖掘”程序认为，10000 MIPS的机器能从杂乱中找到三维物体，并且高分辨率的立体视觉程序用10 MIPS的机器在1小时内演示了这一点。随着计算机速度和内存的增大，数据量问题逐渐不再是问题。
除了纯粹的运算规模，也得考虑其他因素。对1 MIPS的机器，最适用的程序是能有效处理检测数据的手工程序。
12 more annotations...
- 机器人视觉程序想要聪相应图像中得到一个边缘检测或移动检测数据的话，需执行100条计算机指令。一百万次检测则对应执行一亿条指令，1000MIPS的机器1秒能重复十次上述过程，基本上可与人的视网膜相比，新型高端个人计算机刚好能达到这个性能。
- 程序的实现不但取决于机器的速度，而且还受内存的制约。令人惊奇的是，在计算机发展史中，内存和速度比是非常恒定的
- 我们将内存容量除以速度定义为“时间常数”，粗略表示计算机扫描内存一次需要的时间。1兆字节/1 MIPS表示的结果是1秒。若机器的内存不能满足速度要求，这是新机器典型的情况，似乎运算速度很快，但会不必要地局限于小程序。若机器内存相对速度过剩，可以说它的商业可行性到了尽头，虽能运算大程序，但出奇的慢
- 先前已得出一亿MIPS能模仿人的功能这一结论，在这里做比较的话，前面所述的兆字节/MIPS的原则似乎在神经系统中也成立！交互的机器好象是神经系统的外延，遵从同样的时间常数。与其用机器与外界交互，不如让不同速度/存储比的人来完成
- 视听处理器和高性能飞行器等的控制器需要反就迅速，所以每兆字节相应很多MIPS。自动数据库和时变的安全摄像等问题涉及到慢速机器，每个MIPS会对应很多兆字节
- 飞虫似乎比人类反应快好几倍，MIPS的相对量要大一些。动物的情况也是这样，细胞间靠电化学和酶的方式传递信号。尽管植物细胞与动物神经元有本质的区别，一些植物细胞间似乎也能通讯。有一天，我们可能发现植物能记忆许多信息，但处理缓慢
- 二战后，计算机容量每两年翻一番。但进入20世纪80年代，每18个月计算机性能就要翻倍，20世纪90年代后期似乎缩短为12个月
- 通用性的实现是需要代价的。一个通用机器需要十倍或十倍以上的资源来完成一项专用任务。如果任务改变，通用机器可以被重新编程，而专用机器就得被淘汰。
- 以目前的发展速度，类似人的机器人所需要的计算机将在21世纪20年代出现。
- 实现高级计算的关键是微型化，因为小部件的惯性小，在同等能量下运行更快，在给定空间中能集成更多内容
- 集成芯片不但在持续发展，而且在加速前进。制作集成电路的短波长的光被替代，制作中使用了更精确方法来嵌入混杂成份。集成电路的电压在减小，采用了更好的绝缘体、屏蔽方法、散热手段，电路上出现了更有效的晶体管、更密的管腿模式和不辐射的包装材料。如果有钱来刺激的话，总会有办法。事实上，方法在实验室里早有了。因为工程师们当时还没注意这些，他们正在完善已有的技术。当根本问题出现，他们才感到忧虑。当需求的呼声很高时，巨大研究成果会转变为现实的生产。
- 传统晶体管电路依赖于大量电子的流动，当规模很小时，会引起状态波动。但采用单电子晶体管和量子设备可以解决这项难题，它们的工作原理是量子波干涉。这些新设备越小，工作得就越好。干涉图式很适用目前的电路规模，它只用很小能量就可将电子从一边撞到另一边，从而完成操作。因此，这些电路可以在温度很低的环境下工作。这种方法能集成到0.01微米，量子交换在室温下就能进行。

《机器人》第二章－小心！机器车辆！

www.yrobot.net/...71_819.html - Preview

AI book robot

从1948年到1951年，英国心理学家沃尔特（W.Grey Walter）在英国造了六个电子乌龟，它们都带有超微型电子管大脑、旋转光电眼睛、接触开关式触角6。它们能在行走时避开障碍，在有特定光线召唤时能返回出发地。当几只小乌龟在一块时，它们会对彼此间的控制光线和触觉信号有所反应，表现出意想不到的社会行为，例如跳舞。
对于机器而言，计算要比推理容易得多，推理又要比感知和行动容易得多。为什么难易顺序正好与人相反呢？在十亿年的进化过程中，人类祖先的第一个进化选择都为了能在逆境的竞争中赢得生命要素，常常用一生的时间在竞争中更有效地感知和行动。达尔文的进化过程使人类的大脑具有获取和行动的特殊本领───这点因为太平常而常被忽略。
17 more annotations...
- 在20世纪的大部分时间里，每二十年机器计算力就会有上千倍的提高
- 目前，控制最好的商业机器人的计算机能够产生昆虫级的行为。它们的耗资不少，社会用途却很少，仅在工厂作为喷涂车、点焊车、安装电子器件和将东西移来移去
- 在未来十年中，机器全面智能化的进程将大大加速，我相信机器人数量会大增，应用范围会慢慢拓宽以满足普遍要求，最终成为通用机器人
- 移动机器人是这一新生的计算机控制机器人领域中的最年轻一代，并在1985年后成为此领域中的主角。1985年前为数不多的工作重点之一是阿波罗登月计划
- 许多任务似乎只需两个基本的图像操作就能完成，一是挑选出好的特征集（图像中许有特别的区域），另一个是对同一区域从不同视角来寻找。
- 三角测量、障碍检测、机器人运动演绎可以确定三维位置。我寻找实现这些想法的快速算法。采用空间压缩图像技术可完成大量工作，4、16、64以及更多像素的平方被平均为一个数，然后用汇编语言巧妙地编码，而我的算子能在一幅图中挑出许多有用特征，并得到另一幅图。整个过程计算机大约只用10秒钟
- 不对物体的位置进行分类，而是积累位置的“物体性”。尽管特别物体的特性甚至存在与否总被怀疑，可以肯定地认为机器人周围的位置是存在的，能被当作永久性的“存储桶”，以稳定地积累哪怕是内容的一丁点迹象。
- 机器人周围的的区域被分成栅格。程序给栅格的每个单元赋以数值代表积累的证据，也就是说，相应的单元包括一些内容或什么都没有。随着每次声纳激励，含有信息的单元被更新。在回波距离处的单元被占据（因为此处的任何物都将导致更短的回波）。声音强度和检测可靠性随着声纳离中心点的距离增大而减弱，所以得到的证据量也据此发生变化。
- 在运算速度为1/2 MIPS的丹宁机器人上，我们研制了能进行每秒十次声纳反馈信息处理的有效程序。操纵机器人的关键一步，即校正同一区域的两幅映像图，需要3秒钟
- 实验揭示，在狭窄而墙壁光滑的走廊里（恰与我们实验室杂乱环境不同）声波被不断反射，类似光线在充满镜子的大厅里的情况，此时大部分的反馈信息是不正确的，得出的映像图也不会有价值。处理程序是以声纳单元的分类为基础编写的，而此时并不能表明声纳数据真正给出环境信息。
- 环境中独立事件数量的增加可以降低误解概率。增加证据栅格信息的最好途径是，全部用三维处理。在二维处理中，高度不同的门把手和桌面等其他东西被测的总是不一致，可能只是一个模糊映像图。将处理单元的边长减小到10厘米左右，也不会有太大的效果
- 在三维空间中，这些是一致的，不会出现模糊现象，栅格的分辨率会更高，单元内容会更确定。在二维空间中，一把椅子只是占据若干个单元的模糊东西，无法与其它具有同样尺寸的物体区分。在三维空间中，椅子就会有座位、椅腿、椅背，能由形状确认。三维图像可以使程序规划复杂路径，不仅可以四处迂回行进，而且能跨越障碍。视觉出错的可能性变得微乎其微。
- 广角镜覆盖面很广，但它产生明显的“鱼眼”失真。到了1995年，我才编写了克服此不足的程序。程序通过摄像机视察几百个黑点的校准模型，得到一个“纠正”函数，将原始图像校正为一个理想的几何图
- 20世纪80年代后期，状态低迷的“基于模型”（或“基于图像”）机器人人面临着基本问题。在麻省理工学院，精力充沛且善于表演的罗德.布鲁克斯（Rod Brooks）声称，基于模型的方法不切合实际，并示范了如何从没有模型的机器人人获取复杂行为。布鲁克斯的机器被巧妙设计的多反射层或行为集所控制，类似格雷.沃尔特的小乌龟或霍普金斯野兽，但更为复杂些。在行为集中，一种元素可能使机器人避开障碍，而根据检测，另一种元素或许使机器人沿墙行走。有些元素对另一些元素有所响应或无视它们。每个元素其实是适度分享处理器的简单程序，典型地，可以处理许多行为元素的机器人也许有1 MIPS的运算能力。自然界中的昆虫似乎被这种方式所控制，每个动作都由几百个神经元的神经节产生。布鲁克斯的机器人动作很像昆虫，当他的实验室开始研制六条腿行走的微型机器人时更说明了这点
- 但是新奇感过后，人们不得不承认，在完成复杂动作方面，基于行为的机器人可靠性并不比集成多控制器的机器人要好
- 20世纪90年代，基于模型的机器人可以在室内行驶，也可穿越乡村。它们含有一些反射行为集的元素，用于避开障碍和处理险情。也许受此影响，布鲁克斯研制了机器人“Cog”.它被大量可学习的反射行为所控制。迄今为止，它可以通过学习来用视觉跟踪物体，并抓获物体。与我们人类的神经系统很相似，但我认为它的模仿是低层次的。需要足够运算时间的Cog，其方法思路也许有助于追溯人类的进化和发展，或许可以导致全智能机器。虽然如此，我认为模仿高层的抽象性是更快的一条途径，直接可以借鉴有用的计算机科学和工程的成果
- 大部分实用自动机必然是按基于行为的思路而研制的。尽管对可预测性的设计必然会降低逼真度，但用电线连接的机电继电器很像简单昆虫的神经节，五十年来，它们一直用于控制自动电梯和工业机器。从20世纪80年代开始，微处理器。系统复杂性在慢慢增大，但大部分控制器仍以简单的方式对刺激产生反应

《机器人》第一章－逃逸速度

www.yrobot.net/...71_818.html - Preview

AI book robot

04 Nov 07

life.science.humanity: 論生命、智慧與電腦

lifesciencehumanity.blogspot.com/...blog-post_27.html - Preview

AI OnIntelligence

我覺得就某部份而言，短期記憶是一個類似電容的東西，電容可以過濾直流電，也就是只讓變動通過，也可以說，他會麻木。從大腦收到訊號開始，就產生預測，而產生的預測就放置在短期記憶中，只有與短期記憶中不同的東西，才會送到後面的記憶系統
皮質層可能也類似這個樣子，對一個電容，假如你一直充訊號進去，他會一直飽飽的，也就是說他有記憶，而太久沒送訊號進去，因為漏電流的關係，裡面的電會慢慢流失掉，這就是因太久沒接觸而遺忘的原因
2 more annotations...
- 人腦跟電腦還有個不同，電腦有所謂的指令週期，人腦的計算，則比較不像一步步進行的計算，而是一個由記憶系統反映出的過程，人在執行一步步的計算時，其實是在做一系列的記憶反射，所以計算速度是由記憶量以及記憶反射週期決定，記憶反射的速度就由神經元之間傳遞訊息的速度決定
- 現有電腦架構的記憶是擺在計算的上層，人腦的記憶則應該是在計算層的上下都有，並且是相通的。再想想，計算似乎是被記憶包住了，計算其實就是記憶，記憶裡提取不到的東西會被拆解成更小的東西去提取，所以所謂計算層其實是記憶系統的遞迴

16 Sep 07

心靈機器時代 —當電腦超越人腦

www.books.com.tw/...booksfile.php - Preview

AI book brain choice computer

15 Sep 07

機器人：由機器邁向超越人類心智之路

www.books.com.tw/...booksfile.php - Preview

AI book choice robot trend

09 Sep 07

When will computer hardware match the human brain? by Hans Moravec

www.transhumanist.com/...moravec.htm - Preview

5star AI MIPS brain computer megabytes memory trend

The processing power and memory
capacity necessary to match general intellectual performance
of the human brain are estimated. Based on extrapolation of
past trends and on examination of technologies under
development, it is predicted that the required hardware will
be available in cheap machines in the 2020s.
Programs need memory as well as processing speed to do
their work. The ratio of memory to speed has remained
constant during computing history.
15 more annotations...
- Dividing memory
  by speed defines a "time constant," roughly how
  long it takes the computer to run once through its
  memory. One megabyte per MIPS gives one second, a nice
  human interval.
- Machines with less memory for their
  speed, typically new models, seem fast, but unnecessarily
  limited to small programs. Models with more memory for
  their speed, often ones reaching the end of their run,
  can handle larger programs, but unpleasantly slowly.
- Customers maintain the ratio by asking
  "would the next dollar be better spent on more speed
  or more memory?"
- The megabyte/MIPS ratio seems to hold
  for nervous systems too! The contingency is the other way
  around: computers are configured to interact at human
  time scales, and robots interacting with humans seem also
  to be best at that ratio.
- faster
  machines, for instance audio and video processors and
  controllers of high-performance aircraft, have many MIPS
  for each megabyte. Very slow machines, for instance
  time-lapse security cameras and automatic data libraries,
  store many megabytes for each of their MIPS.
- Flying
  insects seem to be a few times faster than humans, so may
  have more MIPS than megabytes. As in animals, cells in
  plants signal one other electrochemically and
  enzymatically. Some plant cells seem specialized for
  communication, though apparently not as extremely as
  animal neurons. One day we may find that plants remember
  much, but process it slowly
- With our conversions, a 100-MIPS robot, for instance
  Navlab, has mental power similar to a 100,000-neuron
  housefly. The following figure rates various entities.
- MIPS and Megabytes.
- Universal computers
  can imitate other entities at their location in the
  diagram, but the more specialized entities cannot. A
  100-million-MIPS computer may be programmed not only to
  think like a human, but also to imitate other
  similarly-sized computers. But humans cannot imitate
  100-million-MIPS computers--our general-purpose
  calculation ability is under a millionth of a MIPS.
- Computers doubled in capacity every two years after the war, a
  pace that became an industry given: companies that wished to grow
  sought to exceed it, companies that failed to keep up lost
  business. In the 1980s the doubling time contracted to 18 months,
  and computer performance in the late 1990s seems to be doubling
  every 12 months.
- Faster than Exponential Growth in
  Computing Power. The number of MIPS in $1000 of
  computer from 1900 to the present. Steady improvements in
  mechanical and electromechanical calculators before World War II
  had increased the speed of calculation a thousandfold over manual
  methods from 1900 to 1940. The pace quickened with the appearance
  of electronic computers during the war, and 1940 to 1980 saw a
  millionfold increase. The pace has been even quicker since then,
  a pace which would make humanlike robots possible before the
  middle of the next century. The vertical scale is logarithmic,
  the major divisions represent thousandfold increases in computer
  performance. Exponential growth would show as a straight line,
  the upward curve indicates faster than exponential growth, or,
  equivalently, an accelerating rate of innovation. The reduced
  spread of the data in the 1990s is probably the result of
  intensified competition: underperforming machines are more
  rapidly squeezed out. The numerical data for this power curve are
  presented in the appendix.
- Chip progress not only continued, it
  sped up. Shorter-wavelength light was substituted, a more precise
  way of implanting impurities was devised, voltages were reduced,
  better insulators, shielding designs, more efficient transistor
  designs, better heat sinks, denser pin patterns and
  non-radioactive packaging materials were found. Where there is
  sufficient financial incentive, there is a way. In fact,
  solutions had been waiting in research labs for years, barely
  noticed by the engineers in the field, who were perfecting
  established processes, and worrying in print as those ran out of
  steam. As the need became acute, enormous resources were
  redirected to draft laboratory possibilities into production
  realities.
- The wave-like nature of matter at very small scales is a problem
  for conventional transistors, which depend on the smooth flow of
  masses of electrons. But, it is a property exploited by a radical
  new class of components known as single-electron transistors and
  quantum dots, which work by the interference of electron waves.
  These new devices work better as they grow smaller. At the scale
  of today's circuits, the interference patterns are so fine that
  it takes only a little heat energy to bump electrons from crest
  to crest, scrambling their operation. Thus, these circuits have
  been demonstrated mostly at a few degrees above absolute zero.
  But, as the devices are reduced, the interference patterns widen,
  and it takes ever larger energy to disrupt them. Scaled to about
  0.01 micrometers, quantum interference switching works at room
  temperature.
- The big freeze. From
  1960 to 1990 the cost of computers used in AI research declined,
  as their numbers dilution absorbed computer-efficiency gains
  during the period, and the power available to individual AI
  programs remained almost unchanged at 1 MIPS, barely insect
  power. AI computer cost bottomed in 1990, and since then power
  has doubled yearly, to several hundred MIPS by 1998. The major
  visible exception is computer chess (shown by a progression of
  knights), whose prestige lured the resources of major computer
  companies and the talents of programmers and machine designers.
  Exceptions also exist in less public competitions, like petroleum
  exploration and intelligence gathering, whose high return on
  investment gave them regular access to the largest computers.
- Agony to ecstasy. In
  forty years, computer chess progressed from the lowest depth to
  the highest peak of human chess performance. It took a handful of
  good ideas, culled by trial and error from a larger number of
  possibilities, an accumulation of previously evaluated game
  openings and endings, good adjustment of position scores, and
  especially a ten-million-fold increase in the number of
  alternative move sequences the machines can explore. Note that
  chess machines reached world champion performance as their
  (specialized) processing power reached about 1/30 human, by our
  brain to computer measure. Since it is plausible that Garry
  Kasparov (but hardly anyone else) can apply his brainpower to the
  problems of chess with an efficiency of 1/30, the result supports
  that retina-based extrapolation. In coming decades, as
  general-purpose computer power grows beyond Deep Blue's
  specialized strength, machines will begin to match humans in more
  common skills.

Robot: mere machine to transcendent mind

www.frc.ri.cmu.edu/...book97 - Preview

AI book choice robot trend

Peek at 1998 Moravec book - Chapter 3. Power and Presence

www.frc.ri.cmu.edu/...index.html - Preview

AI MIPS book computer memory robot trend

Power and Capacity
Information handling capacity in computers has been growing about ten
million times faster than it did in nervous systems during our
evolution. The power doubled every two years in the 1950s, 1960s and
1970s, doubled every 18 months in the 1980s, and is now doubling each
year.
3 more annotations...
- Computer power: 1900-1997
- AI Computer power: 1900-1997
- Progress in Computer Chess, a thin slice of AI

03 Jun 07

Pyro, Python Robotics: Pyro

pyrorobotics.org/ - Preview

AI Python robot simulator

31 May 07

機器人的進化：人工智慧與機器人學的新世紀

www.books.com.tw/...booksfile.php - Preview

AI book choice robot survey

Impossible Minds

www.worldscibooks.com/...p023.html - Preview

5star AI NN book choice probability robot weightless

30 May 07

第一套商业机器人智能软体即将面世

www.roboticfan.com/...391.shtml - Preview

AI iRobot news robot

在Nordin博士的实验室里，机器人已经开始像婴儿一样开始学习爬行，慢慢开始学会平衡以及走路，甚至爬楼梯。而且机器人还开始研究放在它们面前的物体，甚至会归纳出物体的属性，并且开始玩弄它们，就像小孩一样。一个4足机器人在不小心损坏一只脚的情况下，能够慢慢学习如何使用3只脚来保持平衡并且有效的行走。在实验室里，类似的项目还有很多。

York Jong's Library tagged AI → View Popular

Antiobjects - Wikipedia, the free encyclopedia

Introduction to AI Robotics

Flesh and Machines 我們都是機器人

科幻中的非人類

智慧型控制：分析與設計

創智慧-理解人腦運作，打造智慧機器

群智能算法及其应用

《机器人》第三章－能力和显现

《机器人》第二章－小心！机器车辆！

《机器人》第一章－逃逸速度

life.science.humanity: 論生命、智慧與電腦

心靈機器時代 —當電腦超越人腦

機器人：由機器邁向超越人類心智之路

When will computer hardware match the human brain? by Hans Moravec

Robot: mere machine to transcendent mind

Peek at 1998 Moravec book - Chapter 3. Power and Presence

Power and Capacity

Computer power: 1900-1997

AI Computer power: 1900-1997

Progress in Computer Chess, a thin slice of AI

Pyro, Python Robotics: Pyro

機器人的進化：人工智慧與機器人學的新世紀

Impossible Minds

第一套商业机器人智能软体即将面世

Selected Tags

Related Tags

Sponsored Links

Top Contributors

Groups interested in AI

Advanced Concepts Team

Artificial Intelligence Research

Robotics