Mercurial > cortex

annotate thesis/cortex.org @ 448:af13fc73e851

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
completing second part of first chapter.
author Robert McIntyre <rlm@mit.edu>
date Tue, 25 Mar 2014 22:54:41 -0400
parents 284316604be0
children 09b7c8dd4365
rev   line source
rlm@425 1 #+title: =CORTEX=
rlm@425 2 #+author: Robert McIntyre
rlm@425 3 #+email: rlm@mit.edu
rlm@425 4 #+description: Using embodied AI to facilitate Artificial Imagination.
rlm@425 5 #+keywords: AI, clojure, embodiment
rlm@422 6
rlm@437 7
rlm@439 8 * Empathy and Embodiment as problem solving strategies
rlm@437 9
rlm@437 10 By the end of this thesis, you will have seen a novel approach to
rlm@437 11 interpreting video using embodiment and empathy. You will have also
rlm@437 12 seen one way to efficiently implement empathy for embodied
rlm@447 13 creatures. Finally, you will become familiar with =CORTEX=, a system
rlm@447 14 for designing and simulating creatures with rich senses, which you
rlm@447 15 may choose to use in your own research.
rlm@437 16
rlm@441 17 This is the core vision of my thesis: That one of the important ways
rlm@441 18 in which we understand others is by imagining ourselves in their
rlm@441 19 position and emphatically feeling experiences relative to our own
rlm@441 20 bodies. By understanding events in terms of our own previous
rlm@441 21 corporeal experience, we greatly constrain the possibilities of what
rlm@441 22 would otherwise be an unwieldy exponential search. This extra
rlm@441 23 constraint can be the difference between easily understanding what
rlm@441 24 is happening in a video and being completely lost in a sea of
rlm@441 25 incomprehensible color and movement.
rlm@435 26
rlm@436 27 ** Recognizing actions in video is extremely difficult
rlm@437 28
rlm@447 29 Consider for example the problem of determining what is happening
rlm@447 30 in a video of which this is one frame:
rlm@437 31
rlm@441 32 #+caption: A cat drinking some water. Identifying this action is
rlm@441 33 #+caption: beyond the state of the art for computers.
rlm@441 34 #+ATTR_LaTeX: :width 7cm
rlm@441 35 [[./images/cat-drinking.jpg]]
rlm@441 36
rlm@441 37 It is currently impossible for any computer program to reliably
rlm@447 38 label such a video as ``drinking''. And rightly so -- it is a very
rlm@441 39 hard problem! What features can you describe in terms of low level
rlm@441 40 functions of pixels that can even begin to describe at a high level
rlm@441 41 what is happening here?
rlm@437 42
rlm@447 43 Or suppose that you are building a program that recognizes chairs.
rlm@448 44 How could you ``see'' the chair in figure \ref{hidden-chair}?
rlm@441 45
rlm@441 46 #+caption: The chair in this image is quite obvious to humans, but I
rlm@448 47 #+caption: doubt that any modern computer vision program can find it.
rlm@441 48 #+name: hidden-chair
rlm@441 49 #+ATTR_LaTeX: :width 10cm
rlm@441 50 [[./images/fat-person-sitting-at-desk.jpg]]
rlm@441 51
rlm@441 52 Finally, how is it that you can easily tell the difference between
rlm@441 53 how the girls /muscles/ are working in figure \ref{girl}?
rlm@441 54
rlm@441 55 #+caption: The mysterious ``common sense'' appears here as you are able
rlm@441 56 #+caption: to discern the difference in how the girl's arm muscles
rlm@441 57 #+caption: are activated between the two images.
rlm@441 58 #+name: girl
rlm@448 59 #+ATTR_LaTeX: :width 7cm
rlm@441 60 [[./images/wall-push.png]]
rlm@437 61
rlm@441 62 Each of these examples tells us something about what might be going
rlm@441 63 on in our minds as we easily solve these recognition problems.
rlm@441 64
rlm@441 65 The hidden chairs show us that we are strongly triggered by cues
rlm@447 66 relating to the position of human bodies, and that we can determine
rlm@447 67 the overall physical configuration of a human body even if much of
rlm@447 68 that body is occluded.
rlm@437 69
rlm@441 70 The picture of the girl pushing against the wall tells us that we
rlm@441 71 have common sense knowledge about the kinetics of our own bodies.
rlm@441 72 We know well how our muscles would have to work to maintain us in
rlm@441 73 most positions, and we can easily project this self-knowledge to
rlm@441 74 imagined positions triggered by images of the human body.
rlm@441 75
rlm@441 76 ** =EMPATH= neatly solves recognition problems
rlm@441 77
rlm@441 78 I propose a system that can express the types of recognition
rlm@441 79 problems above in a form amenable to computation. It is split into
rlm@441 80 four parts:
rlm@441 81
rlm@448 82 - Free/Guided Play :: The creature moves around and experiences the
rlm@448 83 world through its unique perspective. Many otherwise
rlm@448 84 complicated actions are easily described in the language of a
rlm@448 85 full suite of body-centered, rich senses. For example,
rlm@448 86 drinking is the feeling of water sliding down your throat, and
rlm@448 87 cooling your insides. It's often accompanied by bringing your
rlm@448 88 hand close to your face, or bringing your face close to water.
rlm@448 89 Sitting down is the feeling of bending your knees, activating
rlm@448 90 your quadriceps, then feeling a surface with your bottom and
rlm@448 91 relaxing your legs. These body-centered action descriptions
rlm@448 92 can be either learned or hard coded.
rlm@448 93 - Posture Imitation :: When trying to interpret a video or image,
rlm@448 94 the creature takes a model of itself and aligns it with
rlm@448 95 whatever it sees. This alignment can even cross species, as
rlm@448 96 when humans try to align themselves with things like ponies,
rlm@448 97 dogs, or other humans with a different body type.
rlm@448 98 - Empathy :: The alignment triggers associations with
rlm@448 99 sensory data from prior experiences. For example, the
rlm@448 100 alignment itself easily maps to proprioceptive data. Any
rlm@448 101 sounds or obvious skin contact in the video can to a lesser
rlm@448 102 extent trigger previous experience. Segments of previous
rlm@448 103 experiences are stitched together to form a coherent and
rlm@448 104 complete sensory portrait of the scene.
rlm@448 105 - Recognition :: With the scene described in terms of first
rlm@448 106 person sensory events, the creature can now run its
rlm@447 107 action-identification programs on this synthesized sensory
rlm@447 108 data, just as it would if it were actually experiencing the
rlm@447 109 scene first-hand. If previous experience has been accurately
rlm@447 110 retrieved, and if it is analogous enough to the scene, then
rlm@447 111 the creature will correctly identify the action in the scene.
rlm@447 112
rlm@441 113 For example, I think humans are able to label the cat video as
rlm@447 114 ``drinking'' because they imagine /themselves/ as the cat, and
rlm@441 115 imagine putting their face up against a stream of water and
rlm@441 116 sticking out their tongue. In that imagined world, they can feel
rlm@441 117 the cool water hitting their tongue, and feel the water entering
rlm@447 118 their body, and are able to recognize that /feeling/ as drinking.
rlm@447 119 So, the label of the action is not really in the pixels of the
rlm@447 120 image, but is found clearly in a simulation inspired by those
rlm@447 121 pixels. An imaginative system, having been trained on drinking and
rlm@447 122 non-drinking examples and learning that the most important
rlm@447 123 component of drinking is the feeling of water sliding down one's
rlm@447 124 throat, would analyze a video of a cat drinking in the following
rlm@447 125 manner:
rlm@441 126
rlm@447 127 1. Create a physical model of the video by putting a ``fuzzy''
rlm@447 128 model of its own body in place of the cat. Possibly also create
rlm@447 129 a simulation of the stream of water.
rlm@441 130
rlm@441 131 2. Play out this simulated scene and generate imagined sensory
rlm@441 132 experience. This will include relevant muscle contractions, a
rlm@441 133 close up view of the stream from the cat's perspective, and most
rlm@441 134 importantly, the imagined feeling of water entering the
rlm@443 135 mouth. The imagined sensory experience can come from a
rlm@441 136 simulation of the event, but can also be pattern-matched from
rlm@441 137 previous, similar embodied experience.
rlm@441 138
rlm@441 139 3. The action is now easily identified as drinking by the sense of
rlm@441 140 taste alone. The other senses (such as the tongue moving in and
rlm@441 141 out) help to give plausibility to the simulated action. Note that
rlm@441 142 the sense of vision, while critical in creating the simulation,
rlm@441 143 is not critical for identifying the action from the simulation.
rlm@441 144
rlm@441 145 For the chair examples, the process is even easier:
rlm@441 146
rlm@441 147 1. Align a model of your body to the person in the image.
rlm@441 148
rlm@441 149 2. Generate proprioceptive sensory data from this alignment.
rlm@437 150
rlm@441 151 3. Use the imagined proprioceptive data as a key to lookup related
rlm@441 152 sensory experience associated with that particular proproceptive
rlm@441 153 feeling.
rlm@437 154
rlm@443 155 4. Retrieve the feeling of your bottom resting on a surface, your
rlm@443 156 knees bent, and your leg muscles relaxed.
rlm@437 157
rlm@441 158 5. This sensory information is consistent with the =sitting?=
rlm@441 159 sensory predicate, so you (and the entity in the image) must be
rlm@441 160 sitting.
rlm@440 161
rlm@441 162 6. There must be a chair-like object since you are sitting.
rlm@440 163
rlm@441 164 Empathy offers yet another alternative to the age-old AI
rlm@441 165 representation question: ``What is a chair?'' --- A chair is the
rlm@441 166 feeling of sitting.
rlm@441 167
rlm@441 168 My program, =EMPATH= uses this empathic problem solving technique
rlm@441 169 to interpret the actions of a simple, worm-like creature.
rlm@437 170
rlm@441 171 #+caption: The worm performs many actions during free play such as
rlm@441 172 #+caption: curling, wiggling, and resting.
rlm@441 173 #+name: worm-intro
rlm@446 174 #+ATTR_LaTeX: :width 15cm
rlm@445 175 [[./images/worm-intro-white.png]]
rlm@437 176
rlm@447 177 #+caption: =EMPATH= recognized and classified each of these poses by
rlm@447 178 #+caption: inferring the complete sensory experience from
rlm@447 179 #+caption: proprioceptive data.
rlm@441 180 #+name: worm-recognition-intro
rlm@446 181 #+ATTR_LaTeX: :width 15cm
rlm@445 182 [[./images/worm-poses.png]]
rlm@441 183
rlm@441 184 One powerful advantage of empathic problem solving is that it
rlm@441 185 factors the action recognition problem into two easier problems. To
rlm@441 186 use empathy, you need an /aligner/, which takes the video and a
rlm@441 187 model of your body, and aligns the model with the video. Then, you
rlm@441 188 need a /recognizer/, which uses the aligned model to interpret the
rlm@441 189 action. The power in this method lies in the fact that you describe
rlm@448 190 all actions form a body-centered viewpoint. You are less tied to
rlm@447 191 the particulars of any visual representation of the actions. If you
rlm@441 192 teach the system what ``running'' is, and you have a good enough
rlm@441 193 aligner, the system will from then on be able to recognize running
rlm@441 194 from any point of view, even strange points of view like above or
rlm@441 195 underneath the runner. This is in contrast to action recognition
rlm@448 196 schemes that try to identify actions using a non-embodied approach.
rlm@448 197 If these systems learn about running as viewed from the side, they
rlm@448 198 will not automatically be able to recognize running from any other
rlm@448 199 viewpoint.
rlm@441 200
rlm@441 201 Another powerful advantage is that using the language of multiple
rlm@441 202 body-centered rich senses to describe body-centerd actions offers a
rlm@441 203 massive boost in descriptive capability. Consider how difficult it
rlm@441 204 would be to compose a set of HOG filters to describe the action of
rlm@447 205 a simple worm-creature ``curling'' so that its head touches its
rlm@447 206 tail, and then behold the simplicity of describing thus action in a
rlm@441 207 language designed for the task (listing \ref{grand-circle-intro}):
rlm@441 208
rlm@446 209 #+caption: Body-centerd actions are best expressed in a body-centered
rlm@446 210 #+caption: language. This code detects when the worm has curled into a
rlm@446 211 #+caption: full circle. Imagine how you would replicate this functionality
rlm@446 212 #+caption: using low-level pixel features such as HOG filters!
rlm@446 213 #+name: grand-circle-intro
rlm@446 214 #+begin_listing clojure
rlm@446 215 #+begin_src clojure
rlm@446 216 (defn grand-circle?
rlm@446 217 "Does the worm form a majestic circle (one end touching the other)?"
rlm@446 218 [experiences]
rlm@446 219 (and (curled? experiences)
rlm@446 220 (let [worm-touch (:touch (peek experiences))
rlm@446 221 tail-touch (worm-touch 0)
rlm@446 222 head-touch (worm-touch 4)]
rlm@446 223 (and (< 0.55 (contact worm-segment-bottom-tip tail-touch))
rlm@446 224 (< 0.55 (contact worm-segment-top-tip head-touch))))))
rlm@446 225 #+end_src
rlm@446 226 #+end_listing
rlm@446 227
rlm@435 228
rlm@437 229 ** =CORTEX= is a toolkit for building sensate creatures
rlm@435 230
rlm@448 231 I built =CORTEX= to be a general AI research platform for doing
rlm@448 232 experiments involving multiple rich senses and a wide variety and
rlm@448 233 number of creatures. I intend it to be useful as a library for many
rlm@448 234 more projects than just this one. =CORTEX= was necessary to meet a
rlm@448 235 need among AI researchers at CSAIL and beyond, which is that people
rlm@448 236 often will invent neat ideas that are best expressed in the
rlm@448 237 language of creatures and senses, but in order to explore those
rlm@448 238 ideas they must first build a platform in which they can create
rlm@448 239 simulated creatures with rich senses! There are many ideas that
rlm@448 240 would be simple to execute (such as =EMPATH=), but attached to them
rlm@448 241 is the multi-month effort to make a good creature simulator. Often,
rlm@448 242 that initial investment of time proves to be too much, and the
rlm@448 243 project must make do with a lesser environment.
rlm@435 244
rlm@448 245 =CORTEX= is well suited as an environment for embodied AI research
rlm@448 246 for three reasons:
rlm@448 247
rlm@448 248 - You can create new creatures using Blender, a popular 3D modeling
rlm@448 249 program. Each sense can be specified using special blender nodes
rlm@448 250 with biologically inspired paramaters. You need not write any
rlm@448 251 code to create a creature, and can use a wide library of
rlm@448 252 pre-existing blender models as a base for your own creatures.
rlm@448 253
rlm@448 254 - =CORTEX= implements a wide variety of senses, including touch,
rlm@448 255 proprioception, vision, hearing, and muscle tension. Complicated
rlm@448 256 senses like touch, and vision involve multiple sensory elements
rlm@448 257 embedded in a 2D surface. You have complete control over the
rlm@448 258 distribution of these sensor elements through the use of simple
rlm@448 259 png image files. In particular, =CORTEX= implements more
rlm@448 260 comprehensive hearing than any other creature simulation system
rlm@448 261 available.
rlm@448 262
rlm@448 263 - =CORTEX= supports any number of creatures and any number of
rlm@448 264 senses. Time in =CORTEX= dialates so that the simulated creatures
rlm@448 265 always precieve a perfectly smooth flow of time, regardless of
rlm@448 266 the actual computational load.
rlm@448 267
rlm@448 268 =CORTEX= is built on top of =jMonkeyEngine3=, which is a video game
rlm@448 269 engine designed to create cross-platform 3D desktop games. =CORTEX=
rlm@448 270 is mainly written in clojure, a dialect of =LISP= that runs on the
rlm@448 271 java virtual machine (JVM). The API for creating and simulating
rlm@448 272 creatures is entirely expressed in clojure. Hearing is implemented
rlm@448 273 as a layer of clojure code on top of a layer of java code on top of
rlm@448 274 a layer of =C++= code which implements a modified version of
rlm@448 275 =OpenAL= to support multiple listeners. =CORTEX= is the only
rlm@448 276 simulation environment that I know of that can support multiple
rlm@448 277 entities that can each hear the world from their own perspective.
rlm@448 278 Other senses also require a small layer of Java code. =CORTEX= also
rlm@448 279 uses =bullet=, a physics simulator written in =C=.
rlm@448 280
rlm@448 281 #+caption: Here is the worm from above modeled in Blender, a free
rlm@448 282 #+caption: 3D-modeling program. Senses and joints are described
rlm@448 283 #+caption: using special nodes in Blender.
rlm@448 284 #+name: worm-recognition-intro
rlm@448 285 #+ATTR_LaTeX: :width 12cm
rlm@448 286 [[./images/blender-worm.png]]
rlm@448 287
rlm@448 288 During one test with =CORTEX=, I created 3,000 entities each with
rlm@448 289 their own independent senses and ran them all at only 1/80 real
rlm@448 290 time. In another test, I created a detailed model of my own hand,
rlm@448 291 equipped with a realistic distribution of touch (more sensitive at
rlm@448 292 the fingertips), as well as eyes and ears, and it ran at around 1/4
rlm@448 293 real time.
rlm@448 294
rlm@448 295 #+caption: Here is the worm from above modeled in Blender, a free
rlm@448 296 #+caption: 3D-modeling program. Senses and joints are described
rlm@448 297 #+caption: using special nodes in Blender.
rlm@448 298 #+name: worm-recognition-intro
rlm@448 299 #+ATTR_LaTeX: :width 15cm
rlm@448 300 [[./images/full-hand.png]]
rlm@448 301
rlm@448 302
rlm@448 303
rlm@448 304
rlm@448 305
rlm@437 306 ** Contributions
rlm@435 307
rlm@436 308 * Building =CORTEX=
rlm@435 309
rlm@436 310 ** To explore embodiment, we need a world, body, and senses
rlm@435 311
rlm@436 312 ** Because of Time, simulation is perferable to reality
rlm@435 313
rlm@436 314 ** Video game engines are a great starting point
rlm@435 315
rlm@436 316 ** Bodies are composed of segments connected by joints
rlm@435 317
rlm@436 318 ** Eyes reuse standard video game components
rlm@436 319
rlm@436 320 ** Hearing is hard; =CORTEX= does it right
rlm@436 321
rlm@436 322 ** Touch uses hundreds of hair-like elements
rlm@436 323
rlm@440 324 ** Proprioception is the sense that makes everything ``real''
rlm@436 325
rlm@436 326 ** Muscles are both effectors and sensors
rlm@436 327
rlm@436 328 ** =CORTEX= brings complex creatures to life!
rlm@436 329
rlm@436 330 ** =CORTEX= enables many possiblities for further research
rlm@435 331
rlm@435 332 * Empathy in a simulated worm
rlm@435 333
rlm@436 334 ** Embodiment factors action recognition into managable parts
rlm@435 335
rlm@436 336 ** Action recognition is easy with a full gamut of senses
rlm@435 337
rlm@437 338 ** Digression: bootstrapping touch using free exploration
rlm@435 339
rlm@436 340 ** \Phi-space describes the worm's experiences
rlm@435 341
rlm@436 342 ** Empathy is the process of tracing though \Phi-space
rlm@435 343
rlm@441 344 ** Efficient action recognition with =EMPATH=
rlm@425 345
rlm@432 346 * Contributions
rlm@432 347 - Built =CORTEX=, a comprehensive platform for embodied AI
rlm@432 348 experiments. Has many new features lacking in other systems, such
rlm@432 349 as sound. Easy to model/create new creatures.
rlm@432 350 - created a novel concept for action recognition by using artificial
rlm@432 351 imagination.
rlm@426 352
rlm@436 353 In the second half of the thesis I develop a computational model of
rlm@436 354 empathy, using =CORTEX= as a base. Empathy in this context is the
rlm@436 355 ability to observe another creature and infer what sorts of sensations
rlm@436 356 that creature is feeling. My empathy algorithm involves multiple
rlm@436 357 phases. First is free-play, where the creature moves around and gains
rlm@436 358 sensory experience. From this experience I construct a representation
rlm@447 359 of the creature's sensory state space, which I call \Phi-space. Using
rlm@447 360 \Phi-space, I construct an efficient function for enriching the
rlm@436 361 limited data that comes from observing another creature with a full
rlm@436 362 compliment of imagined sensory data based on previous experience. I
rlm@436 363 can then use the imagined sensory data to recognize what the observed
rlm@436 364 creature is doing and feeling, using straightforward embodied action
rlm@436 365 predicates. This is all demonstrated with using a simple worm-like
rlm@436 366 creature, and recognizing worm-actions based on limited data.
rlm@432 367
rlm@436 368 Embodied representation using multiple senses such as touch,
rlm@436 369 proprioception, and muscle tension turns out be be exceedingly
rlm@436 370 efficient at describing body-centered actions. It is the ``right
rlm@436 371 language for the job''. For example, it takes only around 5 lines of
rlm@436 372 LISP code to describe the action of ``curling'' using embodied
rlm@436 373 primitives. It takes about 8 lines to describe the seemingly
rlm@436 374 complicated action of wiggling.
rlm@432 375
rlm@437 376
rlm@437 377
rlm@437 378 * COMMENT names for cortex
rlm@447 379 - bioland
rlm@447 380
rlm@447 381
rlm@447 382
rlm@447 383
rlm@447 384 # An anatomical joke:
rlm@447 385 # - Training
rlm@447 386 # - Skeletal imitation
rlm@447 387 # - Sensory fleshing-out
rlm@447 388 # - Classification