Hinton's CapsNet explains about entity's pose, lighting and deformation, instead of shape feature and pattern. While I reading Hinton's idea, I thought about Holonomic brain theory.
"If a holographic image, containing the image of an apple, is cut in half and then illuminated by a laser, each half will contain the overall image of the apple! Even if the halves are sliced again and again, from each little piece of the film can be restored the image of a whole apple (although the image will become less clear with the reduction of the parts)."
Our brain is also very similar. We don't store images in brain, we store some transformed code. We don't lose our memory, it just fade out.
Subscribe to:
Post Comments (Atom)
-
Opening the black box of Deep Neural Networks via Information - https://arxiv.org/pdf/1703.00810.pdf 지금까지 딥 러닝은 어떻게 동작하는지 이해할 수 없다고 믿어져왔다...
-
음성 인공지능 분야에서 스타트업이 생각해볼 수 있는 전략은 아마 다음과 같이 3가지 정도가 있을 것이다: 독자적 Vertical 음성 인공지능 Application 구축 기 음성 플랫폼을 활용한 B2B2C 형태의 비지니스 구축 기 음성 플랫폼...
-
As mentioned ago, I've been forming up the Hamburg project with Hyunsik Choi. Let's see more detail in the diagram of computing met...
Las Vegas Resort & Casino, LLC | Mashantucket, CT
ReplyDelete› hotel-and-casino- › hotel-and-casino- 4 광명 출장마사지 days 서귀포 출장샵 ago — 평택 출장안마 4 days ago Las Vegas Resort & Casino, LLC, 광양 출장마사지 is 여수 출장마사지 a wholly owned subsidiary of Boyd Gaming Corporation. The property offers 800,000 square feet of