SlideShare a Scribd company logo
1 of 51
Download to read offline
The Effects of Noisy Labels
Keunwoo.Choi

@qmul.ac.uk
on deep convolutional neural networks for music tagging
arXiv:1706.02361
abstract
1. Introduction
@KeunwooChoi
2014--present: PhD, Queen Mary University of London

2016--present: Buzzmusiq lnc.

2016/ 06--12: Visiting PhD, NYU

2015/ 06--09: Intern, Naver Labs

2011--2014: Audio research team, ETRI

2009--2011: Applied Acoustic Lab, EECS, SNU

2005--2009: EECS, SNU

Papers on ISMIR/ICASSP/IEEE Trans./Etc.

Python/Keras/Pytorch
Contents
• Intro to music x deep learning [15min]

• Tagging dataset and noise [30min]

• Label vector analysis [10min]
The Effects of Noisy Labels
Keunwoo.Choi

@qmul.ac.uk
on deep convolutional neural networks for music tagging
György Fazekas, Kyunghyun Cho, Mark Sandler
arXiv:1706.02361
0. INTRODUCTION TO
MUSIC X DEEP LEARNING
Find out the

SoTA-coolness method

eg) ResNets, GANs, VGG, Attention,
Seq2seq, VAE..
Google if someone applied it on
music
“domain adaptation”

i.e., 

words → notes

image → STFT
Profit!
Examples
Issues
Data efficiency := required amount of data for the
same perception
8-bit x 28x28
vs.
16-bit x 1000000
45 MB
50,000 training items
10,000 testing items
10 categories
quick decoding (.jpg)
vs.
1200 MB
1,000 training + test items
10 categories
slow decoding (.mp3)
“it takes approx. 5x times with the same performance in practice”
End-to-end is not always desirable
Preprocessing audio files
Mp3’s

(30s)
x 1M
650GB
Npy’s
(decoded)
x 1M
300GB
Npy’s
(specgram)
x 1M
300GB
num_FFT?	
window_length?	
hop_length?
Npy’s
(specgram)
x 1M
300GB
Npy’s
(specgram)
x 1M
300GB
• It takes time
• and storage
• while all except one will be deleted
• therefore no one cares about it
Kapre; for quicker audio preprocessing
Copyright
How to get away?
Solutions Notes Examples
Non-copyrighted music Free Music Archieve
FMA-dataset

BSS Eval
Releasing YouTube URL “You break the rule, not us” AudioSet
Crawling the previews Dangerous
Million Song Dataset
Sharing the previews Even more dangerous
The Effects of Noisy Labels
Keunwoo.Choi

@qmul.ac.uk
on deep convolutional neural networks for music tagging
György Fazekas, Kyunghyun Cho, Mark Sandler
arXiv:1706.02361
(A short Q&A? +)
1. INTRODUCTION
Tagging
• Anyone can tag any words (or non-words) to any song

• The quality is ****.

• Poor, innocent, (financially) poor researchers need to use it
Tagging
(Tag, count)
rock 101071
pop 69159
alternative 55777
indie 48175
electronic 46270
female vocalists 42565
favorites 39921
00s 31432
Awesome 26248
american 22694
seen live 20705
cool 19581
Favorite18864
Favourites 17722
female vocalist 17328
guitar 17302
loved 12483
favorite songs 12392
heard on Pandora 10470
USA 8725
2000s 8671
Favourite Songs 8661
drjazzmrfunkmusic 8364
77davez-all-tracks7278
fav 6155
bass 3364
songs I absolutely love
3293
vocals 2369
drums2281
🤔
Female vocalists
Male vocalist
Guitar
Bass
Vocals
Drums
0% 25% 50% 75% 100%
True False
Questions
How noisy?
Is training
alright?
How about
evaluation?
What are
they
learning?
The Effects of Noisy Labels
Keunwoo.Choi

@qmul.ac.uk
on deep convolutional neural networks for music tagging
György Fazekas, Kyunghyun Cho, Mark Sandler
arXiv:1706.02361
2. HOW NOISY?

IS TRAINING OK?
Measuring the noise
• We need strongly-labelled re-annotations
• Instrumentation labels are (sort of) objective

(instrumental, female vocal, male vocal, guitar)
• 242K songs are still a lot → select a subset (or two)!
I can do it!
..but not
all of them
Strongly labelling: Subset100
• Subset100: random 50 from ‘True’ 

+ random 50 from ‘False’ (for each label)
Instrumental
Female vocalists
Male vocalist
Guitar
True False
50songs 50songs
50 50
50 50
50 50
Strongly labelling: Subset400
• Subset400: Just random 400 items
242K songs × 50 tags
400 songs
4 tagsSubset400
🎵🖊.......................😭
AFTER
BEFORE
Evaluating groundtruth on Subset100
0
25
50
75
100
+ Error rate Precision
Instrumental female voc
male vocal guitar
0
25
50
75
100
- Error rate Recall
Instrumental female voc
male vocal guitar
#Occurrences estimation
0
20
40
60
80
In all, by GT My estimation

using S100
My re-annotation

on S400
Instrumental female voc male vocal guitar
Again, with box plots
{Instrumental, female vocalists}
vs.
{male vocalists, guitar}
Group A vs B, but why?
• Tagging ‘vocals’, ‘drums’, ‘bass’ is like.. 

→ They’re not tag-worthy

→ Let’s call it ‘taggability’
Female vocalists
Male vocalist
Guitar
Bass
Vocals
Drums
0% 25% 50% 75% 100
True False
***?
What’s on
the desk?
The hypothesis
If unusual → high taggability.
Instrumental, female vocal :
high taggability
Male vocal, guitar:
low taggability
The hypothesis
If unusual → high taggability.
If high taggability
→ less false negative = higher recall (of GT)
Instrumental, female vocal :
high taggability,
less false neg, higher recall
Male vocal, guitar:
low taggability,

more false neg, lower recall
The hypothesis
If unusual → high taggability.
If high taggability
→ less false negative = higher recall (of GT)
If higher recall (=more reliable GT),
→ ?
[33] Choi et al. 2017, Convolutional recu...
Hypothesis
If unusual → high taggability.
If high taggability
→ less false negative = higher recall (of GT)
If higher recall (=more reliable GT),
→ ?
Performance(AUC)
!!!
The hypothesis
If unusual → high taggability.
If high taggability
→ less false negative = higher recall (of GT)
Instrumental, female vocal :
high taggability,
less false neg, higher recall,
better classification
Male vocal, guitar:
low taggability,

more false neg, lower recall,
worse classification
If higher recall (=more reliable GT),
→ better classification
The Effects of Noisy Labels
Keunwoo.Choi

@qmul.ac.uk
on deep convolutional neural networks for music tagging
György Fazekas, Kyunghyun Cho, Mark Sandler
arXiv:1706.02361
3. IS EVALUATION OK?
Really?
So, we evaluate the classifier based on..
🤔
I need a noise-free groundtruth...
Evaluate the evaluation
242K songs × 50 tags
400 songs
4 tagsSubset400
HAHAHAH!Subset400!
Results
Evaluate the evaluation
Interesting! With such noise, 

the results are still okay.
It’s not perfect though.
HAHAHA!
The Effects of Noisy Labels
Keunwoo.Choi

@qmul.ac.uk
on deep convolutional neural networks for music tagging
György Fazekas, Kyunghyun Cho, Mark Sandler
arXiv:1706.02361
(A short Q&A? +)
4. LABEL VECTOR
ANALYSIS
Label vector
(50,	50)
Label vector similarity
• Similarity between labels

according to the trained convnet.
Label vector
Label vector vs co-occurrence (GT)
Label vector vs co-occurrence (GT)
• Mostly, LV reproduces the groundtruth.

• Except: similar pairs only by label vector:

(sad, beautiful), (happy, catchy), (rnb, sexy)
‘Sad songs are beautiful.’
‘Catchy songs are often happy songs.’
‘R&B claims to be sexy.’
🤔 Makes sense..
The Effects of Noisy Labels
Keunwoo.Choi

@qmul.ac.uk
on deep convolutional neural networks for music tagging
György Fazekas, Kyunghyun Cho, Mark Sandler
arXiv:1706.02361
5. CONCLUSIONS
Conclusions
• We quantified how noisy weakly-labelled groundtruth is.

• We conjectured why some labels are noisier.

• We showed what happens to the noisier labels on training
and evaluation.

• We investigated what a convnet learns.
The Effects of Noisy Labels
Keunwoo.Choi

@qmul.ac.uk
on deep convolutional neural networks for music tagging
György Fazekas, Kyunghyun Cho, Mark Sandler
arXiv:1706.02361
Links
My blog | blog post 1, blog post 2 | Paper!

More Related Content

More from NAVER Engineering

디자인 시스템에 직방 ZUIX
디자인 시스템에 직방 ZUIX디자인 시스템에 직방 ZUIX
디자인 시스템에 직방 ZUIXNAVER Engineering
 
진화하는 디자인 시스템(걸음마 편)
진화하는 디자인 시스템(걸음마 편)진화하는 디자인 시스템(걸음마 편)
진화하는 디자인 시스템(걸음마 편)NAVER Engineering
 
서비스 운영을 위한 디자인시스템 프로젝트
서비스 운영을 위한 디자인시스템 프로젝트서비스 운영을 위한 디자인시스템 프로젝트
서비스 운영을 위한 디자인시스템 프로젝트NAVER Engineering
 
BPL(Banksalad Product Language) 무야호
BPL(Banksalad Product Language) 무야호BPL(Banksalad Product Language) 무야호
BPL(Banksalad Product Language) 무야호NAVER Engineering
 
이번 생에 디자인 시스템은 처음이라
이번 생에 디자인 시스템은 처음이라이번 생에 디자인 시스템은 처음이라
이번 생에 디자인 시스템은 처음이라NAVER Engineering
 
날고 있는 여러 비행기 넘나 들며 정비하기
날고 있는 여러 비행기 넘나 들며 정비하기날고 있는 여러 비행기 넘나 들며 정비하기
날고 있는 여러 비행기 넘나 들며 정비하기NAVER Engineering
 
쏘카프레임 구축 배경과 과정
 쏘카프레임 구축 배경과 과정 쏘카프레임 구축 배경과 과정
쏘카프레임 구축 배경과 과정NAVER Engineering
 
플랫폼 디자이너 없이 디자인 시스템을 구축하는 프로덕트 디자이너의 우당탕탕 고통 연대기
플랫폼 디자이너 없이 디자인 시스템을 구축하는 프로덕트 디자이너의 우당탕탕 고통 연대기플랫폼 디자이너 없이 디자인 시스템을 구축하는 프로덕트 디자이너의 우당탕탕 고통 연대기
플랫폼 디자이너 없이 디자인 시스템을 구축하는 프로덕트 디자이너의 우당탕탕 고통 연대기NAVER Engineering
 
200820 NAVER TECH CONCERT 15_Code Review is Horse(코드리뷰는 말이야)(feat.Latte)
200820 NAVER TECH CONCERT 15_Code Review is Horse(코드리뷰는 말이야)(feat.Latte)200820 NAVER TECH CONCERT 15_Code Review is Horse(코드리뷰는 말이야)(feat.Latte)
200820 NAVER TECH CONCERT 15_Code Review is Horse(코드리뷰는 말이야)(feat.Latte)NAVER Engineering
 
200819 NAVER TECH CONCERT 03_화려한 코루틴이 내 앱을 감싸네! 코루틴으로 작성해보는 깔끔한 비동기 코드
200819 NAVER TECH CONCERT 03_화려한 코루틴이 내 앱을 감싸네! 코루틴으로 작성해보는 깔끔한 비동기 코드200819 NAVER TECH CONCERT 03_화려한 코루틴이 내 앱을 감싸네! 코루틴으로 작성해보는 깔끔한 비동기 코드
200819 NAVER TECH CONCERT 03_화려한 코루틴이 내 앱을 감싸네! 코루틴으로 작성해보는 깔끔한 비동기 코드NAVER Engineering
 
200819 NAVER TECH CONCERT 10_맥북에서도 아이맥프로에서 빌드하는 것처럼 빌드 속도 빠르게 하기
200819 NAVER TECH CONCERT 10_맥북에서도 아이맥프로에서 빌드하는 것처럼 빌드 속도 빠르게 하기200819 NAVER TECH CONCERT 10_맥북에서도 아이맥프로에서 빌드하는 것처럼 빌드 속도 빠르게 하기
200819 NAVER TECH CONCERT 10_맥북에서도 아이맥프로에서 빌드하는 것처럼 빌드 속도 빠르게 하기NAVER Engineering
 
200819 NAVER TECH CONCERT 08_성능을 고민하는 슬기로운 개발자 생활
200819 NAVER TECH CONCERT 08_성능을 고민하는 슬기로운 개발자 생활200819 NAVER TECH CONCERT 08_성능을 고민하는 슬기로운 개발자 생활
200819 NAVER TECH CONCERT 08_성능을 고민하는 슬기로운 개발자 생활NAVER Engineering
 
200819 NAVER TECH CONCERT 05_모르면 손해보는 Android 디버깅/분석 꿀팁 대방출
200819 NAVER TECH CONCERT 05_모르면 손해보는 Android 디버깅/분석 꿀팁 대방출200819 NAVER TECH CONCERT 05_모르면 손해보는 Android 디버깅/분석 꿀팁 대방출
200819 NAVER TECH CONCERT 05_모르면 손해보는 Android 디버깅/분석 꿀팁 대방출NAVER Engineering
 
200819 NAVER TECH CONCERT 09_Case.xcodeproj - 좋은 동료로 거듭나기 위한 노하우
200819 NAVER TECH CONCERT 09_Case.xcodeproj - 좋은 동료로 거듭나기 위한 노하우200819 NAVER TECH CONCERT 09_Case.xcodeproj - 좋은 동료로 거듭나기 위한 노하우
200819 NAVER TECH CONCERT 09_Case.xcodeproj - 좋은 동료로 거듭나기 위한 노하우NAVER Engineering
 
200820 NAVER TECH CONCERT 14_야 너두 할 수 있어. 비전공자, COBOL 개발자를 거쳐 네이버에서 FE 개발하게 된...
200820 NAVER TECH CONCERT 14_야 너두 할 수 있어. 비전공자, COBOL 개발자를 거쳐 네이버에서 FE 개발하게 된...200820 NAVER TECH CONCERT 14_야 너두 할 수 있어. 비전공자, COBOL 개발자를 거쳐 네이버에서 FE 개발하게 된...
200820 NAVER TECH CONCERT 14_야 너두 할 수 있어. 비전공자, COBOL 개발자를 거쳐 네이버에서 FE 개발하게 된...NAVER Engineering
 
200820 NAVER TECH CONCERT 13_네이버에서 오픈 소스 개발을 통해 성장하는 방법
200820 NAVER TECH CONCERT 13_네이버에서 오픈 소스 개발을 통해 성장하는 방법200820 NAVER TECH CONCERT 13_네이버에서 오픈 소스 개발을 통해 성장하는 방법
200820 NAVER TECH CONCERT 13_네이버에서 오픈 소스 개발을 통해 성장하는 방법NAVER Engineering
 
200820 NAVER TECH CONCERT 12_상반기 네이버 인턴을 돌아보며
200820 NAVER TECH CONCERT 12_상반기 네이버 인턴을 돌아보며200820 NAVER TECH CONCERT 12_상반기 네이버 인턴을 돌아보며
200820 NAVER TECH CONCERT 12_상반기 네이버 인턴을 돌아보며NAVER Engineering
 
200820 NAVER TECH CONCERT 11_빠르게 성장하는 슈퍼루키로 거듭나기
200820 NAVER TECH CONCERT 11_빠르게 성장하는 슈퍼루키로 거듭나기200820 NAVER TECH CONCERT 11_빠르게 성장하는 슈퍼루키로 거듭나기
200820 NAVER TECH CONCERT 11_빠르게 성장하는 슈퍼루키로 거듭나기NAVER Engineering
 
200819 NAVER TECH CONCERT 07_신입 iOS 개발자 개발업무 적응기
200819 NAVER TECH CONCERT 07_신입 iOS 개발자 개발업무 적응기200819 NAVER TECH CONCERT 07_신입 iOS 개발자 개발업무 적응기
200819 NAVER TECH CONCERT 07_신입 iOS 개발자 개발업무 적응기NAVER Engineering
 

More from NAVER Engineering (20)

React vac pattern
React vac patternReact vac pattern
React vac pattern
 
디자인 시스템에 직방 ZUIX
디자인 시스템에 직방 ZUIX디자인 시스템에 직방 ZUIX
디자인 시스템에 직방 ZUIX
 
진화하는 디자인 시스템(걸음마 편)
진화하는 디자인 시스템(걸음마 편)진화하는 디자인 시스템(걸음마 편)
진화하는 디자인 시스템(걸음마 편)
 
서비스 운영을 위한 디자인시스템 프로젝트
서비스 운영을 위한 디자인시스템 프로젝트서비스 운영을 위한 디자인시스템 프로젝트
서비스 운영을 위한 디자인시스템 프로젝트
 
BPL(Banksalad Product Language) 무야호
BPL(Banksalad Product Language) 무야호BPL(Banksalad Product Language) 무야호
BPL(Banksalad Product Language) 무야호
 
이번 생에 디자인 시스템은 처음이라
이번 생에 디자인 시스템은 처음이라이번 생에 디자인 시스템은 처음이라
이번 생에 디자인 시스템은 처음이라
 
날고 있는 여러 비행기 넘나 들며 정비하기
날고 있는 여러 비행기 넘나 들며 정비하기날고 있는 여러 비행기 넘나 들며 정비하기
날고 있는 여러 비행기 넘나 들며 정비하기
 
쏘카프레임 구축 배경과 과정
 쏘카프레임 구축 배경과 과정 쏘카프레임 구축 배경과 과정
쏘카프레임 구축 배경과 과정
 
플랫폼 디자이너 없이 디자인 시스템을 구축하는 프로덕트 디자이너의 우당탕탕 고통 연대기
플랫폼 디자이너 없이 디자인 시스템을 구축하는 프로덕트 디자이너의 우당탕탕 고통 연대기플랫폼 디자이너 없이 디자인 시스템을 구축하는 프로덕트 디자이너의 우당탕탕 고통 연대기
플랫폼 디자이너 없이 디자인 시스템을 구축하는 프로덕트 디자이너의 우당탕탕 고통 연대기
 
200820 NAVER TECH CONCERT 15_Code Review is Horse(코드리뷰는 말이야)(feat.Latte)
200820 NAVER TECH CONCERT 15_Code Review is Horse(코드리뷰는 말이야)(feat.Latte)200820 NAVER TECH CONCERT 15_Code Review is Horse(코드리뷰는 말이야)(feat.Latte)
200820 NAVER TECH CONCERT 15_Code Review is Horse(코드리뷰는 말이야)(feat.Latte)
 
200819 NAVER TECH CONCERT 03_화려한 코루틴이 내 앱을 감싸네! 코루틴으로 작성해보는 깔끔한 비동기 코드
200819 NAVER TECH CONCERT 03_화려한 코루틴이 내 앱을 감싸네! 코루틴으로 작성해보는 깔끔한 비동기 코드200819 NAVER TECH CONCERT 03_화려한 코루틴이 내 앱을 감싸네! 코루틴으로 작성해보는 깔끔한 비동기 코드
200819 NAVER TECH CONCERT 03_화려한 코루틴이 내 앱을 감싸네! 코루틴으로 작성해보는 깔끔한 비동기 코드
 
200819 NAVER TECH CONCERT 10_맥북에서도 아이맥프로에서 빌드하는 것처럼 빌드 속도 빠르게 하기
200819 NAVER TECH CONCERT 10_맥북에서도 아이맥프로에서 빌드하는 것처럼 빌드 속도 빠르게 하기200819 NAVER TECH CONCERT 10_맥북에서도 아이맥프로에서 빌드하는 것처럼 빌드 속도 빠르게 하기
200819 NAVER TECH CONCERT 10_맥북에서도 아이맥프로에서 빌드하는 것처럼 빌드 속도 빠르게 하기
 
200819 NAVER TECH CONCERT 08_성능을 고민하는 슬기로운 개발자 생활
200819 NAVER TECH CONCERT 08_성능을 고민하는 슬기로운 개발자 생활200819 NAVER TECH CONCERT 08_성능을 고민하는 슬기로운 개발자 생활
200819 NAVER TECH CONCERT 08_성능을 고민하는 슬기로운 개발자 생활
 
200819 NAVER TECH CONCERT 05_모르면 손해보는 Android 디버깅/분석 꿀팁 대방출
200819 NAVER TECH CONCERT 05_모르면 손해보는 Android 디버깅/분석 꿀팁 대방출200819 NAVER TECH CONCERT 05_모르면 손해보는 Android 디버깅/분석 꿀팁 대방출
200819 NAVER TECH CONCERT 05_모르면 손해보는 Android 디버깅/분석 꿀팁 대방출
 
200819 NAVER TECH CONCERT 09_Case.xcodeproj - 좋은 동료로 거듭나기 위한 노하우
200819 NAVER TECH CONCERT 09_Case.xcodeproj - 좋은 동료로 거듭나기 위한 노하우200819 NAVER TECH CONCERT 09_Case.xcodeproj - 좋은 동료로 거듭나기 위한 노하우
200819 NAVER TECH CONCERT 09_Case.xcodeproj - 좋은 동료로 거듭나기 위한 노하우
 
200820 NAVER TECH CONCERT 14_야 너두 할 수 있어. 비전공자, COBOL 개발자를 거쳐 네이버에서 FE 개발하게 된...
200820 NAVER TECH CONCERT 14_야 너두 할 수 있어. 비전공자, COBOL 개발자를 거쳐 네이버에서 FE 개발하게 된...200820 NAVER TECH CONCERT 14_야 너두 할 수 있어. 비전공자, COBOL 개발자를 거쳐 네이버에서 FE 개발하게 된...
200820 NAVER TECH CONCERT 14_야 너두 할 수 있어. 비전공자, COBOL 개발자를 거쳐 네이버에서 FE 개발하게 된...
 
200820 NAVER TECH CONCERT 13_네이버에서 오픈 소스 개발을 통해 성장하는 방법
200820 NAVER TECH CONCERT 13_네이버에서 오픈 소스 개발을 통해 성장하는 방법200820 NAVER TECH CONCERT 13_네이버에서 오픈 소스 개발을 통해 성장하는 방법
200820 NAVER TECH CONCERT 13_네이버에서 오픈 소스 개발을 통해 성장하는 방법
 
200820 NAVER TECH CONCERT 12_상반기 네이버 인턴을 돌아보며
200820 NAVER TECH CONCERT 12_상반기 네이버 인턴을 돌아보며200820 NAVER TECH CONCERT 12_상반기 네이버 인턴을 돌아보며
200820 NAVER TECH CONCERT 12_상반기 네이버 인턴을 돌아보며
 
200820 NAVER TECH CONCERT 11_빠르게 성장하는 슈퍼루키로 거듭나기
200820 NAVER TECH CONCERT 11_빠르게 성장하는 슈퍼루키로 거듭나기200820 NAVER TECH CONCERT 11_빠르게 성장하는 슈퍼루키로 거듭나기
200820 NAVER TECH CONCERT 11_빠르게 성장하는 슈퍼루키로 거듭나기
 
200819 NAVER TECH CONCERT 07_신입 iOS 개발자 개발업무 적응기
200819 NAVER TECH CONCERT 07_신입 iOS 개발자 개발업무 적응기200819 NAVER TECH CONCERT 07_신입 iOS 개발자 개발업무 적응기
200819 NAVER TECH CONCERT 07_신입 iOS 개발자 개발업무 적응기
 

Recently uploaded

Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Scott Keck-Warren
 
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphSIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphNeo4j
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024Scott Keck-Warren
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 3652toLead Limited
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking MenDelhi Call girls
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...HostedbyConfluent
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhisoniya singh
 
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesSinan KOZAK
 
How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?XfilesPro
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitecturePixlogix Infotech
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountPuma Security, LLC
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
Artificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraArtificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraDeakin University
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):comworks
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsSnow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsHyundai Motor Group
 

Recently uploaded (20)

Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024
 
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food ManufacturingPigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food Manufacturing
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphSIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
 
SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
 
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen Frames
 
How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC Architecture
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
Artificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning eraArtificial intelligence in the post-deep learning era
Artificial intelligence in the post-deep learning era
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter RoadsSnow Chain-Integrated Tire for a Safe Drive on Winter Roads
Snow Chain-Integrated Tire for a Safe Drive on Winter Roads
 

The effects of noisy labels on deep convolutional neural networks for music tagging

  • 1. The Effects of Noisy Labels Keunwoo.Choi @qmul.ac.uk on deep convolutional neural networks for music tagging arXiv:1706.02361
  • 3. @KeunwooChoi 2014--present: PhD, Queen Mary University of London 2016--present: Buzzmusiq lnc. 2016/ 06--12: Visiting PhD, NYU 2015/ 06--09: Intern, Naver Labs 2011--2014: Audio research team, ETRI 2009--2011: Applied Acoustic Lab, EECS, SNU 2005--2009: EECS, SNU Papers on ISMIR/ICASSP/IEEE Trans./Etc. Python/Keras/Pytorch
  • 4. Contents • Intro to music x deep learning [15min] • Tagging dataset and noise [30min] • Label vector analysis [10min]
  • 5. The Effects of Noisy Labels Keunwoo.Choi @qmul.ac.uk on deep convolutional neural networks for music tagging György Fazekas, Kyunghyun Cho, Mark Sandler arXiv:1706.02361 0. INTRODUCTION TO MUSIC X DEEP LEARNING
  • 6. Find out the
 SoTA-coolness method
 eg) ResNets, GANs, VGG, Attention, Seq2seq, VAE.. Google if someone applied it on music “domain adaptation”
 i.e., 
 words → notes
 image → STFT Profit!
  • 9. Data efficiency := required amount of data for the same perception
  • 10. 8-bit x 28x28 vs. 16-bit x 1000000 45 MB 50,000 training items 10,000 testing items 10 categories quick decoding (.jpg) vs. 1200 MB 1,000 training + test items 10 categories slow decoding (.mp3)
  • 11. “it takes approx. 5x times with the same performance in practice” End-to-end is not always desirable
  • 12. Preprocessing audio files Mp3’s
 (30s) x 1M 650GB Npy’s (decoded) x 1M 300GB Npy’s (specgram) x 1M 300GB num_FFT? window_length? hop_length? Npy’s (specgram) x 1M 300GB Npy’s (specgram) x 1M 300GB • It takes time • and storage • while all except one will be deleted • therefore no one cares about it
  • 13. Kapre; for quicker audio preprocessing
  • 14.
  • 16. How to get away? Solutions Notes Examples Non-copyrighted music Free Music Archieve FMA-dataset BSS Eval Releasing YouTube URL “You break the rule, not us” AudioSet Crawling the previews Dangerous Million Song Dataset Sharing the previews Even more dangerous
  • 17. The Effects of Noisy Labels Keunwoo.Choi @qmul.ac.uk on deep convolutional neural networks for music tagging György Fazekas, Kyunghyun Cho, Mark Sandler arXiv:1706.02361 (A short Q&A? +) 1. INTRODUCTION
  • 18. Tagging • Anyone can tag any words (or non-words) to any song • The quality is ****. • Poor, innocent, (financially) poor researchers need to use it
  • 19. Tagging (Tag, count) rock 101071 pop 69159 alternative 55777 indie 48175 electronic 46270 female vocalists 42565 favorites 39921 00s 31432 Awesome 26248 american 22694 seen live 20705 cool 19581 Favorite18864 Favourites 17722 female vocalist 17328 guitar 17302 loved 12483 favorite songs 12392 heard on Pandora 10470 USA 8725 2000s 8671 Favourite Songs 8661 drjazzmrfunkmusic 8364 77davez-all-tracks7278 fav 6155 bass 3364 songs I absolutely love 3293 vocals 2369 drums2281
  • 21. Questions How noisy? Is training alright? How about evaluation? What are they learning?
  • 22. The Effects of Noisy Labels Keunwoo.Choi @qmul.ac.uk on deep convolutional neural networks for music tagging György Fazekas, Kyunghyun Cho, Mark Sandler arXiv:1706.02361 2. HOW NOISY?
 IS TRAINING OK?
  • 23. Measuring the noise • We need strongly-labelled re-annotations • Instrumentation labels are (sort of) objective (instrumental, female vocal, male vocal, guitar) • 242K songs are still a lot → select a subset (or two)! I can do it! ..but not all of them
  • 24. Strongly labelling: Subset100 • Subset100: random 50 from ‘True’ 
 + random 50 from ‘False’ (for each label) Instrumental Female vocalists Male vocalist Guitar True False 50songs 50songs 50 50 50 50 50 50
  • 25. Strongly labelling: Subset400 • Subset400: Just random 400 items 242K songs × 50 tags 400 songs 4 tagsSubset400
  • 27. Evaluating groundtruth on Subset100 0 25 50 75 100 + Error rate Precision Instrumental female voc male vocal guitar 0 25 50 75 100 - Error rate Recall Instrumental female voc male vocal guitar
  • 28. #Occurrences estimation 0 20 40 60 80 In all, by GT My estimation
 using S100 My re-annotation
 on S400 Instrumental female voc male vocal guitar
  • 29. Again, with box plots {Instrumental, female vocalists} vs. {male vocalists, guitar}
  • 30. Group A vs B, but why? • Tagging ‘vocals’, ‘drums’, ‘bass’ is like.. → They’re not tag-worthy → Let’s call it ‘taggability’ Female vocalists Male vocalist Guitar Bass Vocals Drums 0% 25% 50% 75% 100 True False ***? What’s on the desk?
  • 31. The hypothesis If unusual → high taggability. Instrumental, female vocal : high taggability Male vocal, guitar: low taggability
  • 32. The hypothesis If unusual → high taggability. If high taggability → less false negative = higher recall (of GT) Instrumental, female vocal : high taggability, less false neg, higher recall Male vocal, guitar: low taggability,
 more false neg, lower recall
  • 33. The hypothesis If unusual → high taggability. If high taggability → less false negative = higher recall (of GT) If higher recall (=more reliable GT), → ?
  • 34. [33] Choi et al. 2017, Convolutional recu... Hypothesis If unusual → high taggability. If high taggability → less false negative = higher recall (of GT) If higher recall (=more reliable GT), → ? Performance(AUC) !!!
  • 35. The hypothesis If unusual → high taggability. If high taggability → less false negative = higher recall (of GT) Instrumental, female vocal : high taggability, less false neg, higher recall, better classification Male vocal, guitar: low taggability,
 more false neg, lower recall, worse classification If higher recall (=more reliable GT), → better classification
  • 36.
  • 37. The Effects of Noisy Labels Keunwoo.Choi @qmul.ac.uk on deep convolutional neural networks for music tagging György Fazekas, Kyunghyun Cho, Mark Sandler arXiv:1706.02361 3. IS EVALUATION OK?
  • 38. Really? So, we evaluate the classifier based on.. 🤔 I need a noise-free groundtruth...
  • 39. Evaluate the evaluation 242K songs × 50 tags 400 songs 4 tagsSubset400 HAHAHAH!Subset400!
  • 41. Evaluate the evaluation Interesting! With such noise, 
 the results are still okay. It’s not perfect though. HAHAHA!
  • 42. The Effects of Noisy Labels Keunwoo.Choi @qmul.ac.uk on deep convolutional neural networks for music tagging György Fazekas, Kyunghyun Cho, Mark Sandler arXiv:1706.02361 (A short Q&A? +) 4. LABEL VECTOR ANALYSIS
  • 44. Label vector similarity • Similarity between labels
 according to the trained convnet.
  • 46. Label vector vs co-occurrence (GT)
  • 47. Label vector vs co-occurrence (GT) • Mostly, LV reproduces the groundtruth. • Except: similar pairs only by label vector: (sad, beautiful), (happy, catchy), (rnb, sexy) ‘Sad songs are beautiful.’ ‘Catchy songs are often happy songs.’ ‘R&B claims to be sexy.’ 🤔 Makes sense..
  • 48. The Effects of Noisy Labels Keunwoo.Choi @qmul.ac.uk on deep convolutional neural networks for music tagging György Fazekas, Kyunghyun Cho, Mark Sandler arXiv:1706.02361 5. CONCLUSIONS
  • 49. Conclusions • We quantified how noisy weakly-labelled groundtruth is. • We conjectured why some labels are noisier. • We showed what happens to the noisier labels on training and evaluation. • We investigated what a convnet learns.
  • 50. The Effects of Noisy Labels Keunwoo.Choi @qmul.ac.uk on deep convolutional neural networks for music tagging György Fazekas, Kyunghyun Cho, Mark Sandler arXiv:1706.02361
  • 51. Links My blog | blog post 1, blog post 2 | Paper!