1 . 1 Como a Como a Neurocincia Neurocincia inspira a inspira a - - PowerPoint PPT Presentation

▶
1 1
SMART_READER_LITE
LIVE PREVIEW

1 . 1 Como a Como a Neurocincia Neurocincia inspira a inspira a - - PowerPoint PPT Presentation

Nov 20, 2022 263 likes •1.36k views

1 . 1 Como a Como a Neurocincia Neurocincia inspira a inspira a Inteligncia Artificial Inteligncia Artificial Estevo Uyr Estevo Uyr Pardillos Vieira Pardillos Vieira 1 . 2 1 . 3 "What I cannot create, "What

slide-1
SLIDE 1

1 . 1

slide-2
SLIDE 2

Como a Como a

Neurociência Neurociência

inspira a inspira a

Inteligência Artificial Inteligência Artificial

Estevão Uyrá Estevão Uyrá

Pardillos Vieira Pardillos Vieira

1 . 2

slide-3
SLIDE 3

1 . 3

slide-4
SLIDE 4

"What I cannot create, "What I cannot create, I do not understand" I do not understand"

Richard Feynmann

1 . 4

slide-5
SLIDE 5

Overview Overview

Introdução Peças básicas Neurônio Plasticidade Reforço Condicionamento Replay Redes Back-propagation Cortex Visual

1 . 5

slide-6
SLIDE 6

Overview Overview

Introdução Peças básicas Neurônio Plasticidade Reforço Condicionamento Replay Redes Back-propagation Cortex Visual

2 . 1

slide-7
SLIDE 7

2 . 2

slide-8
SLIDE 8

Fonte: [26]

2 . 3

slide-9
SLIDE 9

2 . 4

slide-10
SLIDE 10

Fonte: [2]

2 . 5

slide-11
SLIDE 11

Fonte: [2]

2 . 6

slide-12
SLIDE 12

Falta algo para atingir a Falta algo para atingir a inteligência humana? inteligência humana?

2 . 7

slide-13
SLIDE 13

Modificado de: [3]

Avestruz

Caminhão

Falta. Falta.

+ =

2 . 8

slide-14
SLIDE 14

2 . 9

slide-15
SLIDE 15

Fonte: [6]

2 . 10

slide-16
SLIDE 16

Fonte: [6]

2 . 11

slide-17
SLIDE 17

[7]

Fonte: [6]

2 . 11

slide-18
SLIDE 18

[7]

Fonte: [6]

Lontra

2 . 11

slide-19
SLIDE 19

[8] [7]

Fonte: [6]

Lontra

2 . 11

slide-20
SLIDE 20

[5]

[8] [7]

Fonte: [6]

Lontra

2 . 11

slide-21
SLIDE 21

[5]

[8] [7] [9]

Fonte: [6]

Lontra

2 . 11

slide-22
SLIDE 22

O que falta? O que falta?

2 . 12

slide-23
SLIDE 23

Overview Overview

Introdução Peças básicas Neurônio Plasticidade Reforço Condicionamento Replay Redes Back-propagation Cortex Visual

3 . 1

slide-24
SLIDE 24

Fonte: [1]

3 . 2

slide-25
SLIDE 25

Modificado de: [12]

3 . 3

slide-26
SLIDE 26

“ McCulloch, W. S., & Pitts, W. (1943).

A logical calculus of the ideas immanent in nervous activity.

The bulletin of mathematical biophysics, 5(4), 115-133.

3 . 4

slide-27
SLIDE 27

3 . 5

slide-28
SLIDE 28

3 . 6

slide-29
SLIDE 29

Fonte: [16]

3 . 7

slide-30
SLIDE 30

Overview Overview

Introdução Peças básicas Neurônio Plasticidade Reforço Condicionamento Replay Redes Back-propagation Cortex Visual

4 . 1

slide-31
SLIDE 31

“ Neurons that fire together,

Neurons that fire together, wire together wire together

Donald Hebb, 1949

4 . 2

slide-32
SLIDE 32

4 . 3

slide-33
SLIDE 33

“ Rosenblatt, F. (1958).

The perceptron: a probabilistic model for information storage and organization in the brain.

Psychological review, 65(6), 386.

4 . 4

slide-34
SLIDE 34

4 . 5

slide-35
SLIDE 35

4 . 6

slide-36
SLIDE 36

O objeto é um círculo? O objeto é vermelho? É uma maçã?

4 . 7

slide-37
SLIDE 37

O objeto é um círculo? O objeto é vermelho? É uma maçã?

4 . 8

slide-38
SLIDE 38

https://www.youtube.com/embed/xpJHhHwR4DQ?enablejsapi=1

4 . 9

slide-39
SLIDE 39

“ The Navy revealed the embryo of an

electronic computer today that it expects will be able to walk, talk, see, write, reproduce itself and be conscious of its existence. The New York Times, 1958 [32]

4 . 10

slide-40
SLIDE 40

4 . 11

slide-41
SLIDE 41

https://www.youtube.com/embed/3JjDmFV_YwQ?enablejsapi=1

4 . 12

slide-42
SLIDE 42

Fonte: [10]

4 . 13

slide-43
SLIDE 43

4 . 14

slide-44
SLIDE 44

The first AI winter The first AI winter

4 . 15

slide-45
SLIDE 45

The first AI winter The first AI winter

4 . 15

slide-46
SLIDE 46

Overview Overview

Introdução Peças básicas Neurônio Plasticidade Reforço Condicionamento Replay Redes Back-propagation Cortex Visual

5 . 1

slide-47
SLIDE 47

https://www.youtube.com/embed/LeZZLdEJNxg?enablejsapi=1

5 . 2

slide-48
SLIDE 48

“ The real question is not

The real question is not whether machines think but whether machines think but whether men do. whether men do.

B.F. Skinner

5 . 3

slide-49
SLIDE 49

Fonte: [29]

5 . 4

slide-50
SLIDE 50

5 . 5

slide-51
SLIDE 51

5 . 6

slide-52
SLIDE 52

5 . 7

slide-53
SLIDE 53

https://www.youtube.com/embed/6LAWovT7FSI?enablejsapi=1

5 . 8

slide-54
SLIDE 54

https://www.youtube.com/embed/aNXhyPj-RsM?enablejsapi=1

5 . 9

slide-55
SLIDE 55

Reinforcement Learning Reinforcement Learning

5 . 10

slide-56
SLIDE 56

Reinforcement Learning Reinforcement Learning

5 . 10

slide-57
SLIDE 57

Reinforcement Learning Reinforcement Learning

5 . 10

slide-58
SLIDE 58

Reinforcement Learning Reinforcement Learning

+=

Reescrevendo Aproximação iterativa Reescrevendo

5 . 10

slide-59
SLIDE 59

Reinforcement Learning Reinforcement Learning

+= +=

Reescrevendo Aproximação iterativa Reescrevendo Aproximação

5 . 10

slide-60
SLIDE 60

Reinforcement Learning Reinforcement Learning

+= +=

Reescrevendo Aproximação iterativa Reescrevendo Aproximação

Sinal dopaminérgico

5 . 10

slide-61
SLIDE 61

Dopamine responses Dopamine responses

Fonte: [31] 5 . 11

slide-62
SLIDE 62

Dopamine responses Dopamine responses

Fonte: [31] 5 . 12

slide-63
SLIDE 63

Dopamine responses Dopamine responses

Fonte: [31] 5 . 13

slide-64
SLIDE 64

5 . 14

slide-65
SLIDE 65

5 . 14

slide-66
SLIDE 66

5 . 15

slide-67
SLIDE 67

Dopamine responses Dopamine responses

Fonte: [30] 5 . 16

slide-68
SLIDE 68

Dopamine responses Dopamine responses

Fonte: [31] 5 . 17

slide-69
SLIDE 69

Overview Overview

Introdução Peças básicas Neurônio Plasticidade Reforço Condicionamento Replay Redes Back-propagation Cortex Visual

6 . 1

slide-70
SLIDE 70

Adaptado de [33] 6 . 2

slide-71
SLIDE 71

Adaptado de [33] 6 . 2

slide-72
SLIDE 72

Adaptado de [33] 6 . 3

slide-73
SLIDE 73

Fonte: [27] 6 . 4

slide-74
SLIDE 74

6 . 5

slide-75
SLIDE 75

6 . 6

slide-76
SLIDE 76

Number of episodes Number of episodes Time up

No replay Replay

6 . 7

slide-77
SLIDE 77

Fonte: [28] 6 . 8

slide-78
SLIDE 78

Overview Overview

Introdução Peças básicas Neurônio Plasticidade Reforço Condicionamento Replay Redes Back-propagation Cortex Visual

7 . 1

slide-79
SLIDE 79

“ Rumelhart, D. E., McClelland, J. L., & PDP Research Group. (1986).

Parallel Distributed Processing (PDP): Exploration in the Microstructure of Cognition (Vol. 1).

7 . 2

slide-80
SLIDE 80

Fonte: [10]

7 . 3

slide-81
SLIDE 81

7 . 4

slide-82
SLIDE 82

7 . 5

slide-83
SLIDE 83

7 . 6

slide-84
SLIDE 84

7 . 7

slide-85
SLIDE 85

7 . 8

slide-86
SLIDE 86

pixel 1 pixel N

7 . 9

slide-87
SLIDE 87

Imagem 1 Imagem 2

=

7 . 10

slide-88
SLIDE 88

Overview Overview

Introdução Peças básicas Neurônio Plasticidade Reforço Condicionamento Replay Redes Back-propagation Cortex Visual

8 . 1

slide-89
SLIDE 89

neocognitron convnets

8 . 2

slide-90
SLIDE 90

neocognitron convnets

Fonte: [13]

8 . 3

slide-91
SLIDE 91

Modificado de: [3]

Avestruz

Caminhão

Falta. Falta.

+ =

8 . 4

slide-92
SLIDE 92

Overview Overview

Introdução Peças básicas Neurônio Plasticidade Reforço Condicionamento Replay Redes Back-propagation Cortex Visual

9 . 1

slide-93
SLIDE 93

Agradecimentos Agradecimentos

Gabriela Melo Squad ConvTech @ DataLab Beatriz Albiero Sami Yamouni Mãe

9 . 2

slide-94
SLIDE 94

Referências Referências

ñoz, F., Boya, J., & Alamo, C. (2006). Neuron theory, the cornerstone of neuroscience, on the centenary of the Nobel Prize award to Santiago Ramón y Cajal. Brain research w, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT press, originalmente do wikipedia C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., & Fergus, R. (2013). Intriguing properties of neural networks. arXiv preprint arXiv:1312.6199. ., Engstrom, L., Ilyas, A., & Kwok, K. (2017). Synthesizing robust adversarial examples. arXiv preprint arXiv:1707.07397. tter.com/otter_news/status/1085504638084886534 th blocks - Six Months Old, por Jessica Merz, CC BY-ND 2.0 ( https://creativecommons.org/licenses/by-nd/2.0/) ww.nature.scot/plants-animals-and-fungi/mammals/land-mammals/otter edium.com/taronga-conservation-society-australia/10-things-you-might-not-know-about-the-adorable-otter-ae4c62df34b9 bbble.com/shots/4939061-Swimming-Otter rlrosaen.com/ml/learning-log/2016-05-24/ CC BY-ND 4.0 (https://creativecommons.org/licenses/by-nc/4.0/)

  • druff ; De Roo et al / CC BY-SA 3.0 via Commons)

gacy.owensboro.kctcs.edu/gcaplan/anat/Notes/API Notes K Neurons.htm

  • g.floydhub.com/building-your-first-convnet/
  • R. (2001). The new phrenology: The limits of localizing cognitive processes in the brain. The MIT press.

ww.cambridgebrainsciences.com/more/articles/brain-diagrams-almost-always-face-the-same-direction br.org/2015/03/stop-noise-from-ruining-your-open-office ww.pexels.com/photo/bird-s-eye-view-of-city-buildings-during-sunset-681336/ milimusica.com.br/blog/as-diferencas-entre-o-violao-de-7-cordas-e-o-tradicional/ ww.frontiersin.org/articles/10.3389/fnbeh.2010.00171/full ww.cardiff.ac.uk/cardiff-university-brain-research-imaging-centre/research/themes/cognitive-neuroscience mesclear.com/feynman-mental-models xabay.com/pt/photos/fundo-papel-de-parede-ma%C3%A7%C3%A3-fruto-1655303/ e, Fair use, https://en.wikipedia.org/w/index.php?curid=17594522 ww.companhiadasletras.com.br/autor.php?codigo=00118 uromonitoriaufpa.blogspot.com/2016/10/aula-8-tato-discriminativo.html thub.com/beatrizalbiero/presentations/blob/master/NLPpresentation%20(1).pdf David R., Masami Tatsuno, and Bruce L. McNaughton. "Fast-forward playback of recent memory sequences in prefrontal cortex during sleep." science 318.5853 (2007): 114 ww.diarioliberdade.org/mundo/361-linguaeducacom/50965-o-modelo-did%C3%A1tico-do-ensino-programado,-segundo-b-f-skinner.html[28] Mnih, Volodymyr, et al. "Hu ement learning." Nature518.7540 (2015): 529.

  • nald, and Patricia H. Janak. "Dopamine prediction errors in reward learning and addiction: from theory to neural circuitry." Neuron 88.2 (2015): 247-263.
  • Wolfram. "Predictive reward signal of dopamine neurons." Journal of neurophysiology80.1 (1998): 1-27.

ww.nytimes.com/1958/07/08/archives/new-navy-device-learns-by-doing-psychologist-shows-embryo-of.html rgaret F., Shantanu P. Jadhav, and Loren M. Frank. "Hippocampal replay in the awake state: a potential substrate for memory consolidation and retrieval." Nature neuroscie

9 . 3

slide-95
SLIDE 95

9 . 4

slide-96
SLIDE 96

Message me: estevao.uyra.pv@gmail.com Essa apresentação: slides.com/estevaouyra/tdc_neuro_ai LinkedIn linkedin.com/in/estevao-uyra-pv/

9 . 5

slide-97
SLIDE 97

9 . 6

slide-98
SLIDE 98

Overview Overview

Introdução Peças básicas Neurônio Plasticidade Reforço Condicionamento Replay Redes Back-propagation Cortex Visual O Corpo

10 . 1

slide-99
SLIDE 99

10 . 2

slide-100
SLIDE 100

[18] [17] [25]

10 . 3

slide-101
SLIDE 101

[18] [17] [25]

10 . 4

slide-102
SLIDE 102

https://www.youtube.com/embed/g0TaYhjpOfo?enablejsapi=1

10 . 5

slide-103
SLIDE 103

We are not thinking machines We are not thinking machines that feel; rather,we are feeling that feel; rather,we are feeling machines that think machines that think

Antonio Damasio

10 . 6

slide-104
SLIDE 104

Agradecimentos Agradecimentos

Gabriela Melo Squad ConvTech @ DataLab Beatriz Albiero Sami Yamouni Mãe

11 . 1

slide-105
SLIDE 105

Referências Referências

ñoz, F., Boya, J., & Alamo, C. (2006). Neuron theory, the cornerstone of neuroscience, on the centenary of the Nobel Prize award to Santiago Ramón y Cajal. Brain research w, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT press, originalmente do wikipedia C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., & Fergus, R. (2013). Intriguing properties of neural networks. arXiv preprint arXiv:1312.6199. ., Engstrom, L., Ilyas, A., & Kwok, K. (2017). Synthesizing robust adversarial examples. arXiv preprint arXiv:1707.07397. tter.com/otter_news/status/1085504638084886534 th blocks - Six Months Old, por Jessica Merz, CC BY-ND 2.0 ( https://creativecommons.org/licenses/by-nd/2.0/) ww.nature.scot/plants-animals-and-fungi/mammals/land-mammals/otter edium.com/taronga-conservation-society-australia/10-things-you-might-not-know-about-the-adorable-otter-ae4c62df34b9 bbble.com/shots/4939061-Swimming-Otter rlrosaen.com/ml/learning-log/2016-05-24/ CC BY-ND 4.0 (https://creativecommons.org/licenses/by-nc/4.0/)

  • druff ; De Roo et al / CC BY-SA 3.0 via Commons)

gacy.owensboro.kctcs.edu/gcaplan/anat/Notes/API Notes K Neurons.htm

  • g.floydhub.com/building-your-first-convnet/
  • R. (2001). The new phrenology: The limits of localizing cognitive processes in the brain. The MIT press.

ww.cambridgebrainsciences.com/more/articles/brain-diagrams-almost-always-face-the-same-direction br.org/2015/03/stop-noise-from-ruining-your-open-office ww.pexels.com/photo/bird-s-eye-view-of-city-buildings-during-sunset-681336/ milimusica.com.br/blog/as-diferencas-entre-o-violao-de-7-cordas-e-o-tradicional/ ww.frontiersin.org/articles/10.3389/fnbeh.2010.00171/full ww.cardiff.ac.uk/cardiff-university-brain-research-imaging-centre/research/themes/cognitive-neuroscience mesclear.com/feynman-mental-models xabay.com/pt/photos/fundo-papel-de-parede-ma%C3%A7%C3%A3-fruto-1655303/ e, Fair use, https://en.wikipedia.org/w/index.php?curid=17594522 ww.companhiadasletras.com.br/autor.php?codigo=00118 uromonitoriaufpa.blogspot.com/2016/10/aula-8-tato-discriminativo.html thub.com/beatrizalbiero/presentations/blob/master/NLPpresentation%20(1).pdf David R., Masami Tatsuno, and Bruce L. McNaughton. "Fast-forward playback of recent memory sequences in prefrontal cortex during sleep." science 318.5853 (2007): 114 ww.diarioliberdade.org/mundo/361-linguaeducacom/50965-o-modelo-did%C3%A1tico-do-ensino-programado,-segundo-b-f-skinner.html[28] Mnih, Volodymyr, et al. "Hu ement learning." Nature518.7540 (2015): 529.

  • nald, and Patricia H. Janak. "Dopamine prediction errors in reward learning and addiction: from theory to neural circuitry." Neuron 88.2 (2015): 247-263.
  • Wolfram. "Predictive reward signal of dopamine neurons." Journal of neurophysiology80.1 (1998): 1-27.

ww.nytimes.com/1958/07/08/archives/new-navy-device-learns-by-doing-psychologist-shows-embryo-of.html rgaret F., Shantanu P. Jadhav, and Loren M. Frank. "Hippocampal replay in the awake state: a potential substrate for memory consolidation and retrieval." Nature neuroscie

11 . 2

slide-106
SLIDE 106

11 . 3

slide-107
SLIDE 107

Message me: estevao.uyra.pv@gmail.com Essa apresentação: slides.com/estevaouyra/tdc_neuro_ai LinkedIn linkedin.com/in/estevao-uyra-pv/

11 . 4

slide-108
SLIDE 108

11 . 5