SlideShare a Scribd company logo
Active Learning for Acoustic
Classification AND
Power Aware Feature Selection for
Audio/Sound Scene
Mulu W. Adhana
/Groep T Campus
Outline
2
• PhD Topic: Active Learning for Acoustic Scene/Event
Classification
o Active Learning: Introduction, principles, …
o Experimental Results
o Current Activities: adaptive, transfer learning, …
• Power Aware Feature Selection
o Proposed Method
o Experiments
o Conclusions
Active Learning: Introduction
3
• Traditional machine learning Techniques (Supervised)
o Abundant labeled data
o Labeling: labour intensive costly and error prone
o Domain expert
• Semi-supervised learning
• Active learning
o Seed information (very small in size)
o Sample unlabeled data points for annotation
o Involves user to actively annotate data point
• Reserved for informative/ambiguous ones
Active Learning Scheme
4
Active Learning in Practice
5
?
automatically
assigned
?
Small distance to separating
hyperplane
Ask the user
Active Learning: Sampling for annotation
6
• How ambiguous?
o Distance from separating line
o Probability assigned to the data point
o Entropy of the probability assigned to the data point
Active Learning: KLR
7
• KLR: alternative to SVM with probabilistic outcome
o Natural extension to multi-class problem
o Probabilistic outcome (𝑝𝑖 = 𝒀 = 𝑦𝑖 𝑿 = 𝑥 ,
𝑖 = 1, … , 𝐶)
o Put threshold on some uncertainty measure: e.g.
𝑝 𝐶 − 𝑝 𝐶−1 > 𝑇 [2]
Results: KLR Active Learning
8
NAR dataset:
• 21 classes
• 431 annotated
examples
• 10-fold cross-
validation
• Reduce 84%
manual annotation
Active, Transfer and Incremental Learning
9
Annotated dataset
from different domain
Small Annotated
dataset
Source-task
• From different env’t
• With different
sensing devices
Knowledge
Target-task
• Slightly different
from the source
task
Power Aware Feature Selection
10
• Wearable Sensor Networks draw energy from battery
o Sensing, processing raw data, packet size,
o bit-depth while acquiring sound scene/event, …
• On-node processing. e.g. movement monitoring
o Extracting expensive features => Energy depletion
• Select inexpensive and information carrying features
Proposed Approach
11
• Apply two criteria: Error rate and cost of extracting a
feature/subset of features(CPU-time)
• Incorporating dependency of feature extraction process
k<<n
Proposed Approach…
12
• Sequential forward selection, wrapper approach:
o Let 𝑌 holds already selected features, then
𝑆 = 𝑆 ∪ min
𝑋 𝑖
(𝑒𝑟𝑟𝑜𝑟(𝑆 ∪ 𝑋𝑖) + 𝜆 ∗ 𝐶𝑜𝑠𝑡(𝑆 ∪ 𝑋𝑖)/(|𝑆 ∪ 𝑋𝑖|)
where 𝜆 tradeoff ,
𝑆 ∪ 𝑋𝑖 −feature cardinality
𝑋𝑖 − random candidate feature
o Stopping criterion: add feature 𝑋𝑖 to Y such that
𝐶 𝑌 ∪ 𝑋𝑖 > 𝐶 𝑌
Feature Extraction: Dependency Graph
13
Experimental Setup
14
• Feature selection on the NAR dataset recorded by
humanoid robot NAO
Classes Audio Features
• 21 classes (door open,
door close, fridge open
close, moving chair, …)
• 431 examples (Sound
Scene)
35 (14-MFCC, 12-GTCC,
pitch Cross-correlation,
Spectral Rolloff, Spectral
Flatness, Spectral Flux,
Spectral Kurtosis , Spectral
Skewness , Spectral Slope,
Spectral Spread and ZCR)
Results
15
Questions

More Related Content

Similar to Active Learning for Acoustic Classification AND Power Aware Feature Selection for Audio/Sound Scene

Lecture 11 - KNN and Clustering, a lecture in subject module Statistical & Ma...
Lecture 11 - KNN and Clustering, a lecture in subject module Statistical & Ma...Lecture 11 - KNN and Clustering, a lecture in subject module Statistical & Ma...
Lecture 11 - KNN and Clustering, a lecture in subject module Statistical & Ma...
Maninda Edirisooriya
 
l1_introduction.pdf
l1_introduction.pdfl1_introduction.pdf
l1_introduction.pdf
Dumith Jayathilaka
 
Energy Monitoring With Self-taught Deep Network
Energy Monitoring With Self-taught Deep NetworkEnergy Monitoring With Self-taught Deep Network
Energy Monitoring With Self-taught Deep Network
Yiqun Hu
 
Toward Rich, User-Defined Aggregation & Subset-Selection Services
Toward Rich, User-Defined Aggregation & Subset-Selection ServicesToward Rich, User-Defined Aggregation & Subset-Selection Services
Toward Rich, User-Defined Aggregation & Subset-Selection Services
The HDF-EOS Tools and Information Center
 
Semantics in Sensor Networks
Semantics in Sensor NetworksSemantics in Sensor Networks
Semantics in Sensor Networks
Oscar Corcho
 
DutchMLSchool 2022 - History and Developments in ML
DutchMLSchool 2022 - History and Developments in MLDutchMLSchool 2022 - History and Developments in ML
DutchMLSchool 2022 - History and Developments in ML
BigML, Inc
 
Training machine learning knn 2017
Training machine learning knn 2017Training machine learning knn 2017
Training machine learning knn 2017
Iwan Sofana
 
K-Nearest Neighbor Classifier
K-Nearest Neighbor ClassifierK-Nearest Neighbor Classifier
K-Nearest Neighbor Classifier
Neha Kulkarni
 
Applications of ann_in_microwave_engineering
Applications of ann_in_microwave_engineeringApplications of ann_in_microwave_engineering
Applications of ann_in_microwave_engineering
prasadhegdegn
 
Data wrangling week 10
Data wrangling week 10Data wrangling week 10
Data wrangling week 10
Ferdin Joe John Joseph PhD
 
Sara Hooker & Sean McPherson, Delta Analytics, at MLconf Seattle 2017
Sara Hooker & Sean McPherson, Delta Analytics, at MLconf Seattle 2017Sara Hooker & Sean McPherson, Delta Analytics, at MLconf Seattle 2017
Sara Hooker & Sean McPherson, Delta Analytics, at MLconf Seattle 2017
MLconf
 
Density based clustering
Density based clusteringDensity based clustering
Density based clustering
YaswanthHariKumarVud
 
Garuda Robotics x DataScience SG Meetup (Sep 2015)
Garuda Robotics x DataScience SG Meetup (Sep 2015)Garuda Robotics x DataScience SG Meetup (Sep 2015)
Garuda Robotics x DataScience SG Meetup (Sep 2015)
Eugene Yan Ziyou
 
background.pptx
background.pptxbackground.pptx
background.pptx
KabileshCm
 
03-06-ACA-Input-FeatureLearning.pdf
03-06-ACA-Input-FeatureLearning.pdf03-06-ACA-Input-FeatureLearning.pdf
03-06-ACA-Input-FeatureLearning.pdf
AlexanderLerch4
 
5.1 mining data streams
5.1 mining data streams5.1 mining data streams
5.1 mining data streams
Krish_ver2
 
Gradient Based Power Line Insulator Detection
Gradient Based Power Line Insulator DetectionGradient Based Power Line Insulator Detection
Gradient Based Power Line Insulator Detection
MD RAIHAN
 
ETA Prediction with Graph Neural Networks in Google Maps
ETA Prediction with Graph Neural Networks in Google MapsETA Prediction with Graph Neural Networks in Google Maps
ETA Prediction with Graph Neural Networks in Google Maps
ivaderivader
 
Clustering: A Scikit Learn Tutorial
Clustering: A Scikit Learn TutorialClustering: A Scikit Learn Tutorial
Clustering: A Scikit Learn Tutorial
Damian R. Mingle, MBA
 
Outlier analysis for Temporal Datasets
Outlier analysis for Temporal DatasetsOutlier analysis for Temporal Datasets
Outlier analysis for Temporal Datasets
QuantUniversity
 

Similar to Active Learning for Acoustic Classification AND Power Aware Feature Selection for Audio/Sound Scene (20)

Lecture 11 - KNN and Clustering, a lecture in subject module Statistical & Ma...
Lecture 11 - KNN and Clustering, a lecture in subject module Statistical & Ma...Lecture 11 - KNN and Clustering, a lecture in subject module Statistical & Ma...
Lecture 11 - KNN and Clustering, a lecture in subject module Statistical & Ma...
 
l1_introduction.pdf
l1_introduction.pdfl1_introduction.pdf
l1_introduction.pdf
 
Energy Monitoring With Self-taught Deep Network
Energy Monitoring With Self-taught Deep NetworkEnergy Monitoring With Self-taught Deep Network
Energy Monitoring With Self-taught Deep Network
 
Toward Rich, User-Defined Aggregation & Subset-Selection Services
Toward Rich, User-Defined Aggregation & Subset-Selection ServicesToward Rich, User-Defined Aggregation & Subset-Selection Services
Toward Rich, User-Defined Aggregation & Subset-Selection Services
 
Semantics in Sensor Networks
Semantics in Sensor NetworksSemantics in Sensor Networks
Semantics in Sensor Networks
 
DutchMLSchool 2022 - History and Developments in ML
DutchMLSchool 2022 - History and Developments in MLDutchMLSchool 2022 - History and Developments in ML
DutchMLSchool 2022 - History and Developments in ML
 
Training machine learning knn 2017
Training machine learning knn 2017Training machine learning knn 2017
Training machine learning knn 2017
 
K-Nearest Neighbor Classifier
K-Nearest Neighbor ClassifierK-Nearest Neighbor Classifier
K-Nearest Neighbor Classifier
 
Applications of ann_in_microwave_engineering
Applications of ann_in_microwave_engineeringApplications of ann_in_microwave_engineering
Applications of ann_in_microwave_engineering
 
Data wrangling week 10
Data wrangling week 10Data wrangling week 10
Data wrangling week 10
 
Sara Hooker & Sean McPherson, Delta Analytics, at MLconf Seattle 2017
Sara Hooker & Sean McPherson, Delta Analytics, at MLconf Seattle 2017Sara Hooker & Sean McPherson, Delta Analytics, at MLconf Seattle 2017
Sara Hooker & Sean McPherson, Delta Analytics, at MLconf Seattle 2017
 
Density based clustering
Density based clusteringDensity based clustering
Density based clustering
 
Garuda Robotics x DataScience SG Meetup (Sep 2015)
Garuda Robotics x DataScience SG Meetup (Sep 2015)Garuda Robotics x DataScience SG Meetup (Sep 2015)
Garuda Robotics x DataScience SG Meetup (Sep 2015)
 
background.pptx
background.pptxbackground.pptx
background.pptx
 
03-06-ACA-Input-FeatureLearning.pdf
03-06-ACA-Input-FeatureLearning.pdf03-06-ACA-Input-FeatureLearning.pdf
03-06-ACA-Input-FeatureLearning.pdf
 
5.1 mining data streams
5.1 mining data streams5.1 mining data streams
5.1 mining data streams
 
Gradient Based Power Line Insulator Detection
Gradient Based Power Line Insulator DetectionGradient Based Power Line Insulator Detection
Gradient Based Power Line Insulator Detection
 
ETA Prediction with Graph Neural Networks in Google Maps
ETA Prediction with Graph Neural Networks in Google MapsETA Prediction with Graph Neural Networks in Google Maps
ETA Prediction with Graph Neural Networks in Google Maps
 
Clustering: A Scikit Learn Tutorial
Clustering: A Scikit Learn TutorialClustering: A Scikit Learn Tutorial
Clustering: A Scikit Learn Tutorial
 
Outlier analysis for Temporal Datasets
Outlier analysis for Temporal DatasetsOutlier analysis for Temporal Datasets
Outlier analysis for Temporal Datasets
 

Recently uploaded

5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
ihlasbinance2003
 
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressionsKuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
Victor Morales
 
Casting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdfCasting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdf
zubairahmad848137
 
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODELDEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
gerogepatton
 
ISPM 15 Heat Treated Wood Stamps and why your shipping must have one
ISPM 15 Heat Treated Wood Stamps and why your shipping must have oneISPM 15 Heat Treated Wood Stamps and why your shipping must have one
ISPM 15 Heat Treated Wood Stamps and why your shipping must have one
Las Vegas Warehouse
 
Manufacturing Process of molasses based distillery ppt.pptx
Manufacturing Process of molasses based distillery ppt.pptxManufacturing Process of molasses based distillery ppt.pptx
Manufacturing Process of molasses based distillery ppt.pptx
Madan Karki
 
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Sinan KOZAK
 
Engine Lubrication performance System.pdf
Engine Lubrication performance System.pdfEngine Lubrication performance System.pdf
Engine Lubrication performance System.pdf
mamamaam477
 
A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...
nooriasukmaningtyas
 
官方认证美国密歇根州立大学毕业证学位证书原版一模一样
官方认证美国密歇根州立大学毕业证学位证书原版一模一样官方认证美国密歇根州立大学毕业证学位证书原版一模一样
官方认证美国密歇根州立大学毕业证学位证书原版一模一样
171ticu
 
basic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdfbasic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdf
NidhalKahouli2
 
CSM Cloud Service Management Presentarion
CSM Cloud Service Management PresentarionCSM Cloud Service Management Presentarion
CSM Cloud Service Management Presentarion
rpskprasana
 
Question paper of renewable energy sources
Question paper of renewable energy sourcesQuestion paper of renewable energy sources
Question paper of renewable energy sources
mahammadsalmanmech
 
22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt
KrishnaveniKrishnara1
 
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEMTIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
HODECEDSIET
 
Computational Engineering IITH Presentation
Computational Engineering IITH PresentationComputational Engineering IITH Presentation
Computational Engineering IITH Presentation
co23btech11018
 
Understanding Inductive Bias in Machine Learning
Understanding Inductive Bias in Machine LearningUnderstanding Inductive Bias in Machine Learning
Understanding Inductive Bias in Machine Learning
SUTEJAS
 
International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...
gerogepatton
 
Embedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoringEmbedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoring
IJECEIAES
 
Recycled Concrete Aggregate in Construction Part III
Recycled Concrete Aggregate in Construction Part IIIRecycled Concrete Aggregate in Construction Part III
Recycled Concrete Aggregate in Construction Part III
Aditya Rajan Patra
 

Recently uploaded (20)

5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
5214-1693458878915-Unit 6 2023 to 2024 academic year assignment (AutoRecovere...
 
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressionsKuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressions
 
Casting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdfCasting-Defect-inSlab continuous casting.pdf
Casting-Defect-inSlab continuous casting.pdf
 
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODELDEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODEL
 
ISPM 15 Heat Treated Wood Stamps and why your shipping must have one
ISPM 15 Heat Treated Wood Stamps and why your shipping must have oneISPM 15 Heat Treated Wood Stamps and why your shipping must have one
ISPM 15 Heat Treated Wood Stamps and why your shipping must have one
 
Manufacturing Process of molasses based distillery ppt.pptx
Manufacturing Process of molasses based distillery ppt.pptxManufacturing Process of molasses based distillery ppt.pptx
Manufacturing Process of molasses based distillery ppt.pptx
 
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
Optimizing Gradle Builds - Gradle DPE Tour Berlin 2024
 
Engine Lubrication performance System.pdf
Engine Lubrication performance System.pdfEngine Lubrication performance System.pdf
Engine Lubrication performance System.pdf
 
A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...A review on techniques and modelling methodologies used for checking electrom...
A review on techniques and modelling methodologies used for checking electrom...
 
官方认证美国密歇根州立大学毕业证学位证书原版一模一样
官方认证美国密歇根州立大学毕业证学位证书原版一模一样官方认证美国密歇根州立大学毕业证学位证书原版一模一样
官方认证美国密歇根州立大学毕业证学位证书原版一模一样
 
basic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdfbasic-wireline-operations-course-mahmoud-f-radwan.pdf
basic-wireline-operations-course-mahmoud-f-radwan.pdf
 
CSM Cloud Service Management Presentarion
CSM Cloud Service Management PresentarionCSM Cloud Service Management Presentarion
CSM Cloud Service Management Presentarion
 
Question paper of renewable energy sources
Question paper of renewable energy sourcesQuestion paper of renewable energy sources
Question paper of renewable energy sources
 
22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt22CYT12-Unit-V-E Waste and its Management.ppt
22CYT12-Unit-V-E Waste and its Management.ppt
 
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEMTIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
TIME DIVISION MULTIPLEXING TECHNIQUE FOR COMMUNICATION SYSTEM
 
Computational Engineering IITH Presentation
Computational Engineering IITH PresentationComputational Engineering IITH Presentation
Computational Engineering IITH Presentation
 
Understanding Inductive Bias in Machine Learning
Understanding Inductive Bias in Machine LearningUnderstanding Inductive Bias in Machine Learning
Understanding Inductive Bias in Machine Learning
 
International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...International Conference on NLP, Artificial Intelligence, Machine Learning an...
International Conference on NLP, Artificial Intelligence, Machine Learning an...
 
Embedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoringEmbedded machine learning-based road conditions and driving behavior monitoring
Embedded machine learning-based road conditions and driving behavior monitoring
 
Recycled Concrete Aggregate in Construction Part III
Recycled Concrete Aggregate in Construction Part IIIRecycled Concrete Aggregate in Construction Part III
Recycled Concrete Aggregate in Construction Part III
 

Active Learning for Acoustic Classification AND Power Aware Feature Selection for Audio/Sound Scene

  • 1. Active Learning for Acoustic Classification AND Power Aware Feature Selection for Audio/Sound Scene Mulu W. Adhana /Groep T Campus
  • 2. Outline 2 • PhD Topic: Active Learning for Acoustic Scene/Event Classification o Active Learning: Introduction, principles, … o Experimental Results o Current Activities: adaptive, transfer learning, … • Power Aware Feature Selection o Proposed Method o Experiments o Conclusions
  • 3. Active Learning: Introduction 3 • Traditional machine learning Techniques (Supervised) o Abundant labeled data o Labeling: labour intensive costly and error prone o Domain expert • Semi-supervised learning • Active learning o Seed information (very small in size) o Sample unlabeled data points for annotation o Involves user to actively annotate data point • Reserved for informative/ambiguous ones
  • 5. Active Learning in Practice 5 ? automatically assigned ? Small distance to separating hyperplane Ask the user
  • 6. Active Learning: Sampling for annotation 6 • How ambiguous? o Distance from separating line o Probability assigned to the data point o Entropy of the probability assigned to the data point
  • 7. Active Learning: KLR 7 • KLR: alternative to SVM with probabilistic outcome o Natural extension to multi-class problem o Probabilistic outcome (𝑝𝑖 = 𝒀 = 𝑦𝑖 𝑿 = 𝑥 , 𝑖 = 1, … , 𝐶) o Put threshold on some uncertainty measure: e.g. 𝑝 𝐶 − 𝑝 𝐶−1 > 𝑇 [2]
  • 8. Results: KLR Active Learning 8 NAR dataset: • 21 classes • 431 annotated examples • 10-fold cross- validation • Reduce 84% manual annotation
  • 9. Active, Transfer and Incremental Learning 9 Annotated dataset from different domain Small Annotated dataset Source-task • From different env’t • With different sensing devices Knowledge Target-task • Slightly different from the source task
  • 10. Power Aware Feature Selection 10 • Wearable Sensor Networks draw energy from battery o Sensing, processing raw data, packet size, o bit-depth while acquiring sound scene/event, … • On-node processing. e.g. movement monitoring o Extracting expensive features => Energy depletion • Select inexpensive and information carrying features
  • 11. Proposed Approach 11 • Apply two criteria: Error rate and cost of extracting a feature/subset of features(CPU-time) • Incorporating dependency of feature extraction process k<<n
  • 12. Proposed Approach… 12 • Sequential forward selection, wrapper approach: o Let 𝑌 holds already selected features, then 𝑆 = 𝑆 ∪ min 𝑋 𝑖 (𝑒𝑟𝑟𝑜𝑟(𝑆 ∪ 𝑋𝑖) + 𝜆 ∗ 𝐶𝑜𝑠𝑡(𝑆 ∪ 𝑋𝑖)/(|𝑆 ∪ 𝑋𝑖|) where 𝜆 tradeoff , 𝑆 ∪ 𝑋𝑖 −feature cardinality 𝑋𝑖 − random candidate feature o Stopping criterion: add feature 𝑋𝑖 to Y such that 𝐶 𝑌 ∪ 𝑋𝑖 > 𝐶 𝑌
  • 14. Experimental Setup 14 • Feature selection on the NAR dataset recorded by humanoid robot NAO Classes Audio Features • 21 classes (door open, door close, fridge open close, moving chair, …) • 431 examples (Sound Scene) 35 (14-MFCC, 12-GTCC, pitch Cross-correlation, Spectral Rolloff, Spectral Flatness, Spectral Flux, Spectral Kurtosis , Spectral Skewness , Spectral Slope, Spectral Spread and ZCR)

Editor's Notes

  1. [1] Karsmakers, Peter, Kristiaan Pelckmans, and Johan AK Suykens. "Multi-class kernel logistic regression: a fixed-size implementation." Neural Networks, 2007. IJCNN 2007. International Joint Conference on. IEEE, 2007. [2] Adhana, Mulu Weldegebreal, Bart Vanrumste, and Peter Karsmakers. "Active Learning for Audio-based Home Monitoring." Proceedings of Benelearn 2016. No. Epub ahead of print. 2016.