The document discusses using machine learning and artificial intelligence to create computational creativity and generate music. It provides examples of using supervised machine learning to understand intent during chats and using neural networks to generate text like Harry Potter fan fiction or music. Specifically, it outlines IBM's Watson and the Watson Beat project which uses reinforcement learning and deep belief networks to compose music by learning from feedback on its guesses.
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Using Computational Creativity to Compose Music
1. Anna Chaney | @BeepDJ (music & AI) / @anna_seg (warning: much Beto love at this handle)
A P P L I E D A R T I F I C I A L I N T E L L I G E N C E : U S I N G C O M P U T A T I O N A L
C R E A T I V I T Y T O C R E A T E M U S I C
#GHC18
2. PAGE 2 | GRACE HOPPER CELEBRATION 2018
PRESENTED BY ANITAB.ORG AND THE ASSOCIATION FOR COMPUTING MACHINERY
#GHC18
#GHC 18
MACHINE LEARNING:
THE MATHEMATICAL
APPROACH TO AI
! " = $
" = !%&
$
Weights ObservablesPredictions
3. PAGE 3 | GRACE HOPPER CELEBRATION 2018
PRESENTED BY ANITAB.ORG AND THE ASSOCIATION FOR COMPUTING MACHINERY #GHC 18
THE DAY JOB:
IBM WATSON
The “Cloud”
The IBM Cloud
WATSON
APIs
4. PAGE 4 | GRACE HOPPER CELEBRATION 2018
PRESENTED BY ANITAB.ORG AND THE ASSOCIATION FOR COMPUTING MACHINERY #GHC 18
ANALOGY: SUPERVISED MACHINE LEARNING
Weights
Observables
Predictions Understanding someone’s intent during a chat
A Transform function (SVM, CNN, multi-layer perceptron, …)
The text typed in by a human while interacting with a virtual agent
5. PAGE 5 | GRACE HOPPER CELEBRATION 2018
PRESENTED BY ANITAB.ORG AND THE ASSOCIATION FOR COMPUTING MACHINERY #GHC 18
ANALOGY: SUPERVISED MACHINE LEARNING
Weights
Observables
Predictions Understanding someone’s intent during a chat
A Transform function (SVM, CNN, multi-layer perceptron, …)
The text typed in by a human while interacting with a virtual agent
Needs examples of human decisions so the proper transform
function can be chosen
6. PAGE 6 | GRACE HOPPER CELEBRATION 2018
PRESENTED BY ANITAB.ORG AND THE ASSOCIATION FOR COMPUTING MACHINERY
#GHC 18
PREDICTIONS CAN BE ANYTHING!!
Harry Potter “Fan Fiction”*
The chapter is not entirely penned by AI. While the
algorithm made suggestions, the decision to choose
the next word in the sentence from a list of
suggested words was with the humans. This can be
seen as an exercise where AI assisted textual content
can be created by human beings for entertainment
RNN used to write Game of Thrones ”Fan Fiction”
• https://github.com/zackthoutt/got-book-6
*source : https://www.analyticsindiamag.com/ai-writes-harry-potter-chapter-predictive-text-algorithm/
Image from
http://botnik.org/content/harry-
potter.html
7. #GHC18
U S I N G M A C H I N E L E A R N I N G T O W R I T E M U S I C
https://magenta.tensorflow.org/
ibm.biz/thewatsonbeat
8. PAGE 8 | GRACE HOPPER CELEBRATION 2018
PRESENTED BY ANITAB.ORG AND THE ASSOCIATION FOR COMPUTING MACHINERY
#GHC18
#GHC 18
TWB: REINFORCEMENT LEARNING
Based on
codified rules
of western
music theory
Composer learns
by guessing then
getting feedback
9. PAGE 9 | GRACE HOPPER CELEBRATION 2018
PRESENTED BY ANITAB.ORG AND THE ASSOCIATION FOR COMPUTING MACHINERY
#GHC18
#GHC 18
TWB: DEEP BELIEF NETWORK
Tone and Duration
both in and out