SlideShare a Scribd company logo
1 of 14
ML Robustness in VEDLIoT
António Casimiro
University of Lisbon
HiPEAC 2022
Budapest, 20 June 2022
2
 Local monitoring of input data correctness
 Check characteristics of input features (general data)
 Build ECDF of training data
 Compare with ECDF of input data
 Find outliers/drifts in data values (time series data)
 Store input data (time series values)
 Forecast next expected input value
 Compare forecast value with received input value
Robustness and safety service
 Remote monitoring of model correctness
 Replicated execution using same input on trusted model
(not for time series data)
 Periodically send input/output data to remote node
 Run trusted model with received input data
 Compare outputs of local and remote models
3
 Aware of a context change
 Measure statistical distance between
 Training data distribution
 Real-world data distribution
 If huge difference in distributions -> Do not trust ML
output
Domain monitoring using statistical
methods
 Calculate empirical cumulative distribution
function (ECDF)
1. Ordering all unique observations in data sample
2. Calculating cumulative probability for each as number of
observations less than or equal to a given observation
divided by the total number of observations:
4
 In practice
 Take 30 images from training set and 30 images from
real-world
 Do that for a specific class images based of classifier
prediction
 For each image take the first pixel in the left corner
 Considered one RGB color channel at time
 Calculate first ECDF on the values of the 30 pixels in the
left corner from training set
 The same for real-world image -> second ECDF
Statistical distance between data
distributions
 Apply statistical distance and save the value
 To that for all the pixels and for the three color channel
 Average distances per color channel and compare the
distances with a given threshold
 Limitations
 Too constrained by the color
 Differences in brightness can fool the method
 Images should be well aligned for proper comparison
5
 Specific models for different environmental conditions
 Can they perform better than a single generic one?
 Experiment with CNN object detector (Yolo4)
 Two different conditions: daytime and night
 Driving dataset with 10 classes (pedestrian, rider, car,
…)
Domain adaptation through split approach
 Training three models:
 Daytime model, only objects during daytime
 Night model, only objects during night
 Daytime and night model, both previous ones
 Testing models:
 On daytime images only
 On night images only
Daytime + Night
Daytime Night
Daytime images: 27,967 Night images: 27,967
Daytime + night images: 55,934
Night images: 3,929
Daytime images: 3,929
Training
Testing
6
 Object detector predicts:
 Location of object (coordinates)
 Class (e.g., a dog)
 Confidence score (a value from 0 to 1)
 Confidence score measures the confidence on:
 Localization
 How likely the box contains an object
 How accurate is the box -> IoU
 Classification
 Precision
 «When it guesses how often does it guess correctly?»
 Recall
 «Has it guessed every time that it should have guessed?»
Performance metric
 Confidence threshold
 Positive detection if confidence score > threshold
 Strict threshold -> less recall
 Precision-recall (PR) curve
 Shows trade-off precision/recall for varying threshold
 mean Average precision (mAP)
 Summarize such plot for all classes
 mAP@0.5 at IoU threshold of 50%
7
 Day and night model performs slightly better in both
conditions
 Trained twice images than the other two
 It “saw” more objects during training
 Better a single model and train as much data as
possible
 Then monitor the output
Results
 Other ways to improve robustness and ensure
correctness of output
 Adversarial training
 Uncertainty quantification
 Explainability methods
 Statistical distance between data distributions
Performance on mAP@0.5
Day test set Night test set
Day model 48.59%
Night model 47.82%
Day and night model 50.58% 49.90%
8
Input monitoring (time series data)
 Framework with offline part (model training) and online
part (error detection in input data)
 Detected errors (due to sensor or communication
faults):
 Omissions
 Outliers
 Drifts
Self
Neighbors
Self + Neighbors
Generated Models
Offline
Online
9
Self
Neighbors
Self + Neighbors
Generated Models
Offline
Training phase
 Multiple MLP models are trained to forecast the next
data value to be received by a target sensor:
 A model using only past data from the target sensor
 A model using data from the target sensor and from
neighbor sensors (if they provide correlated data)
 A model using only data from neighbor sensors (if
they provide correlated data)
 This approach allows to distinguish real events (with
impact on multiple sensors) from outliers (affecting
only the target sensor)
10
Online
Online error detection
 Sensor values go through the monitoring service
 Omissions are detected using timers configured for periodic data
 Sensor data (from multiple sensors) is stored and properly aligned to feed each model input
 Using multiple forecasts (from running the multiple models), outliers can be detected
 The service can also replace outliers for quality assurance
11
Missing Data
Outliers
Corrected Data
Forecasts
Results
12
 Monitoring output through explainability
 Provide more evidence to assess output correctness
 Explainability methods are used to evaluate contribution of each input pixel to the output
 Monitoring model correctness
 Periodically send images to redundant remote trustworthy model for comparison of results
 Safety requirements on the architectural framework
 Safety is one of the clusters of concern addressed at several levels of abstration
Other ongoing work
13
Conclusion
Robustness and safety are important concerns, being addressed from several perspectives
• Monitoring methods for input and output data quality
• Monitoring methods for checking model correctness
• Architecting for safety
Questions?

More Related Content

Similar to HiPEAC2022_António Casimiro presentation

Learning from Computer Simulation to Tackle Real-World Problems
Learning from Computer Simulation to Tackle Real-World ProblemsLearning from Computer Simulation to Tackle Real-World Problems
Learning from Computer Simulation to Tackle Real-World ProblemsNAVER Engineering
 
MEME – An Integrated Tool For Advanced Computational Experiments
MEME – An Integrated Tool For Advanced Computational ExperimentsMEME – An Integrated Tool For Advanced Computational Experiments
MEME – An Integrated Tool For Advanced Computational ExperimentsGIScRG
 
AIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTIONAIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTIONIRJET Journal
 
Input Data Collection and Analysis.pptx
Input Data Collection and Analysis.pptxInput Data Collection and Analysis.pptx
Input Data Collection and Analysis.pptxbitf20m550SenirJusti
 
Comparative Study of Pre-Trained Neural Network Models in Detection of Glaucoma
Comparative Study of Pre-Trained Neural Network Models in Detection of GlaucomaComparative Study of Pre-Trained Neural Network Models in Detection of Glaucoma
Comparative Study of Pre-Trained Neural Network Models in Detection of GlaucomaIRJET Journal
 
Machine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdfMachine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdfCarlos Paredes
 
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...csandit
 
A survey of fault prediction using machine learning algorithms
A survey of fault prediction using machine learning algorithmsA survey of fault prediction using machine learning algorithms
A survey of fault prediction using machine learning algorithmsAhmed Magdy Ezzeldin, MSc.
 
2cee Master Cocomo20071
2cee Master Cocomo200712cee Master Cocomo20071
2cee Master Cocomo20071CS, NcState
 
Wang midterm-defence
Wang midterm-defenceWang midterm-defence
Wang midterm-defenceZhipeng Wang
 
One-Sample Face Recognition Using HMM Model of Fiducial Areas
One-Sample Face Recognition Using HMM Model of Fiducial AreasOne-Sample Face Recognition Using HMM Model of Fiducial Areas
One-Sample Face Recognition Using HMM Model of Fiducial AreasCSCJournals
 
2013: Prototype-based learning and adaptive distances for classification
2013: Prototype-based learning and adaptive distances for classification2013: Prototype-based learning and adaptive distances for classification
2013: Prototype-based learning and adaptive distances for classificationUniversity of Groningen
 
Visual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learningVisual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learningBenjamin Bengfort
 
“Person Re-Identification and Tracking at the Edge: Challenges and Techniques...
“Person Re-Identification and Tracking at the Edge: Challenges and Techniques...“Person Re-Identification and Tracking at the Edge: Challenges and Techniques...
“Person Re-Identification and Tracking at the Edge: Challenges and Techniques...Edge AI and Vision Alliance
 
A Review of Lie Detection Techniques
A Review of Lie Detection TechniquesA Review of Lie Detection Techniques
A Review of Lie Detection TechniquesIRJET Journal
 
A Review of Lie Detection Techniques.pdf
A Review of Lie Detection Techniques.pdfA Review of Lie Detection Techniques.pdf
A Review of Lie Detection Techniques.pdfWhitney Anderson
 
NSL KDD Cup 99 dataset Anomaly Detection using Machine Learning Technique
NSL KDD Cup 99 dataset Anomaly Detection using Machine Learning Technique NSL KDD Cup 99 dataset Anomaly Detection using Machine Learning Technique
NSL KDD Cup 99 dataset Anomaly Detection using Machine Learning Technique Sujeet Suryawanshi
 
Integrated Hidden Markov Model and Kalman Filter for Online Object Tracking
Integrated Hidden Markov Model and Kalman Filter for Online Object TrackingIntegrated Hidden Markov Model and Kalman Filter for Online Object Tracking
Integrated Hidden Markov Model and Kalman Filter for Online Object Trackingijsrd.com
 

Similar to HiPEAC2022_António Casimiro presentation (20)

Ajila (1)
Ajila (1)Ajila (1)
Ajila (1)
 
Learning from Computer Simulation to Tackle Real-World Problems
Learning from Computer Simulation to Tackle Real-World ProblemsLearning from Computer Simulation to Tackle Real-World Problems
Learning from Computer Simulation to Tackle Real-World Problems
 
MEME – An Integrated Tool For Advanced Computational Experiments
MEME – An Integrated Tool For Advanced Computational ExperimentsMEME – An Integrated Tool For Advanced Computational Experiments
MEME – An Integrated Tool For Advanced Computational Experiments
 
AIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTIONAIRLINE FARE PRICE PREDICTION
AIRLINE FARE PRICE PREDICTION
 
Input Data Collection and Analysis.pptx
Input Data Collection and Analysis.pptxInput Data Collection and Analysis.pptx
Input Data Collection and Analysis.pptx
 
1025 track1 Malin
1025 track1 Malin1025 track1 Malin
1025 track1 Malin
 
Comparative Study of Pre-Trained Neural Network Models in Detection of Glaucoma
Comparative Study of Pre-Trained Neural Network Models in Detection of GlaucomaComparative Study of Pre-Trained Neural Network Models in Detection of Glaucoma
Comparative Study of Pre-Trained Neural Network Models in Detection of Glaucoma
 
Machine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdfMachine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdf
 
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
EFFICIENT USE OF HYBRID ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM COMBINED WITH N...
 
A survey of fault prediction using machine learning algorithms
A survey of fault prediction using machine learning algorithmsA survey of fault prediction using machine learning algorithms
A survey of fault prediction using machine learning algorithms
 
2cee Master Cocomo20071
2cee Master Cocomo200712cee Master Cocomo20071
2cee Master Cocomo20071
 
Wang midterm-defence
Wang midterm-defenceWang midterm-defence
Wang midterm-defence
 
One-Sample Face Recognition Using HMM Model of Fiducial Areas
One-Sample Face Recognition Using HMM Model of Fiducial AreasOne-Sample Face Recognition Using HMM Model of Fiducial Areas
One-Sample Face Recognition Using HMM Model of Fiducial Areas
 
2013: Prototype-based learning and adaptive distances for classification
2013: Prototype-based learning and adaptive distances for classification2013: Prototype-based learning and adaptive distances for classification
2013: Prototype-based learning and adaptive distances for classification
 
Visual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learningVisual diagnostics for more effective machine learning
Visual diagnostics for more effective machine learning
 
“Person Re-Identification and Tracking at the Edge: Challenges and Techniques...
“Person Re-Identification and Tracking at the Edge: Challenges and Techniques...“Person Re-Identification and Tracking at the Edge: Challenges and Techniques...
“Person Re-Identification and Tracking at the Edge: Challenges and Techniques...
 
A Review of Lie Detection Techniques
A Review of Lie Detection TechniquesA Review of Lie Detection Techniques
A Review of Lie Detection Techniques
 
A Review of Lie Detection Techniques.pdf
A Review of Lie Detection Techniques.pdfA Review of Lie Detection Techniques.pdf
A Review of Lie Detection Techniques.pdf
 
NSL KDD Cup 99 dataset Anomaly Detection using Machine Learning Technique
NSL KDD Cup 99 dataset Anomaly Detection using Machine Learning Technique NSL KDD Cup 99 dataset Anomaly Detection using Machine Learning Technique
NSL KDD Cup 99 dataset Anomaly Detection using Machine Learning Technique
 
Integrated Hidden Markov Model and Kalman Filter for Online Object Tracking
Integrated Hidden Markov Model and Kalman Filter for Online Object TrackingIntegrated Hidden Markov Model and Kalman Filter for Online Object Tracking
Integrated Hidden Markov Model and Kalman Filter for Online Object Tracking
 

More from VEDLIoT Project

IoT Tech Expo 2023_Micha vor dem Berge presentation
IoT Tech Expo 2023_Micha vor dem Berge presentationIoT Tech Expo 2023_Micha vor dem Berge presentation
IoT Tech Expo 2023_Micha vor dem Berge presentationVEDLIoT Project
 
Computing Frontiers 2023_Pedro Trancoso presentation
Computing Frontiers 2023_Pedro Trancoso presentationComputing Frontiers 2023_Pedro Trancoso presentation
Computing Frontiers 2023_Pedro Trancoso presentationVEDLIoT Project
 
HiPEAC-CSW 2022_Pedro Trancoso presentation
HiPEAC-CSW 2022_Pedro Trancoso presentationHiPEAC-CSW 2022_Pedro Trancoso presentation
HiPEAC-CSW 2022_Pedro Trancoso presentationVEDLIoT Project
 
IoT Week 2022-NGIoT session_Micha vor dem Berge presentation
IoT Week 2022-NGIoT session_Micha vor dem Berge presentationIoT Week 2022-NGIoT session_Micha vor dem Berge presentation
IoT Week 2022-NGIoT session_Micha vor dem Berge presentationVEDLIoT Project
 
Next Generation IoT Architectures_Hans Salomonsson
Next Generation IoT Architectures_Hans SalomonssonNext Generation IoT Architectures_Hans Salomonsson
Next Generation IoT Architectures_Hans SalomonssonVEDLIoT Project
 
CONASENSE 2022_Jens Hagemeyer presentation
CONASENSE 2022_Jens Hagemeyer presentationCONASENSE 2022_Jens Hagemeyer presentation
CONASENSE 2022_Jens Hagemeyer presentationVEDLIoT Project
 
NGIoT standardisation workshops_Jens Hagemeyer presentation
NGIoT standardisation workshops_Jens Hagemeyer presentationNGIoT standardisation workshops_Jens Hagemeyer presentation
NGIoT standardisation workshops_Jens Hagemeyer presentationVEDLIoT Project
 
IoT Tech Expo 2023_Pedro Trancoso presentation
IoT Tech Expo 2023_Pedro Trancoso presentationIoT Tech Expo 2023_Pedro Trancoso presentation
IoT Tech Expo 2023_Pedro Trancoso presentationVEDLIoT Project
 
HiPEAC-CSW 2022_Kevin Mika presentation
HiPEAC-CSW 2022_Kevin Mika presentationHiPEAC-CSW 2022_Kevin Mika presentation
HiPEAC-CSW 2022_Kevin Mika presentationVEDLIoT Project
 
HiPEAC 2022-DL4IoT workshop_René Griessl presentation
HiPEAC 2022-DL4IoT workshop_René Griessl presentationHiPEAC 2022-DL4IoT workshop_René Griessl presentation
HiPEAC 2022-DL4IoT workshop_René Griessl presentationVEDLIoT Project
 
SS-CPSIoT 2023_Kevin Mika and Piotr Zierhoffer presentation
SS-CPSIoT 2023_Kevin Mika and Piotr Zierhoffer presentationSS-CPSIoT 2023_Kevin Mika and Piotr Zierhoffer presentation
SS-CPSIoT 2023_Kevin Mika and Piotr Zierhoffer presentationVEDLIoT Project
 
HiPEAC2023-DL4IoT Workshop_Jean Hagemeyer presentation
HiPEAC2023-DL4IoT Workshop_Jean Hagemeyer presentationHiPEAC2023-DL4IoT Workshop_Jean Hagemeyer presentation
HiPEAC2023-DL4IoT Workshop_Jean Hagemeyer presentationVEDLIoT Project
 
IoT Week 2021_Jens Hagemeyer presentation
IoT Week 2021_Jens Hagemeyer presentationIoT Week 2021_Jens Hagemeyer presentation
IoT Week 2021_Jens Hagemeyer presentationVEDLIoT Project
 
HiPEAC 2022_Marcelo Pasin presentation
HiPEAC 2022_Marcelo Pasin presentationHiPEAC 2022_Marcelo Pasin presentation
HiPEAC 2022_Marcelo Pasin presentationVEDLIoT Project
 
IoT Tech Expo 2023_Marcelo Pasin presentation
IoT Tech Expo 2023_Marcelo Pasin presentationIoT Tech Expo 2023_Marcelo Pasin presentation
IoT Tech Expo 2023_Marcelo Pasin presentationVEDLIoT Project
 
IoT Tech Expo 2023_Hans-Martin Heyn presentation
IoT Tech Expo 2023_Hans-Martin Heyn presentationIoT Tech Expo 2023_Hans-Martin Heyn presentation
IoT Tech Expo 2023_Hans-Martin Heyn presentationVEDLIoT Project
 
HiPEAC 2022_Marco Tassemeier presentation
HiPEAC 2022_Marco Tassemeier presentationHiPEAC 2022_Marco Tassemeier presentation
HiPEAC 2022_Marco Tassemeier presentationVEDLIoT Project
 
HiPEAC Computing Systems Week 2022_Mario Porrmann presentation
HiPEAC Computing Systems Week 2022_Mario Porrmann presentationHiPEAC Computing Systems Week 2022_Mario Porrmann presentation
HiPEAC Computing Systems Week 2022_Mario Porrmann presentationVEDLIoT Project
 
NGIoT Sustainability Workshop 2023_ Hans-Martin Heyn presentation
NGIoT Sustainability Workshop 2023_ Hans-Martin Heyn presentationNGIoT Sustainability Workshop 2023_ Hans-Martin Heyn presentation
NGIoT Sustainability Workshop 2023_ Hans-Martin Heyn presentationVEDLIoT Project
 
EU-IoT Training Workshops Series: AIoT and Edge Machine Learning 2021_Jens Ha...
EU-IoT Training Workshops Series: AIoT and Edge Machine Learning 2021_Jens Ha...EU-IoT Training Workshops Series: AIoT and Edge Machine Learning 2021_Jens Ha...
EU-IoT Training Workshops Series: AIoT and Edge Machine Learning 2021_Jens Ha...VEDLIoT Project
 

More from VEDLIoT Project (20)

IoT Tech Expo 2023_Micha vor dem Berge presentation
IoT Tech Expo 2023_Micha vor dem Berge presentationIoT Tech Expo 2023_Micha vor dem Berge presentation
IoT Tech Expo 2023_Micha vor dem Berge presentation
 
Computing Frontiers 2023_Pedro Trancoso presentation
Computing Frontiers 2023_Pedro Trancoso presentationComputing Frontiers 2023_Pedro Trancoso presentation
Computing Frontiers 2023_Pedro Trancoso presentation
 
HiPEAC-CSW 2022_Pedro Trancoso presentation
HiPEAC-CSW 2022_Pedro Trancoso presentationHiPEAC-CSW 2022_Pedro Trancoso presentation
HiPEAC-CSW 2022_Pedro Trancoso presentation
 
IoT Week 2022-NGIoT session_Micha vor dem Berge presentation
IoT Week 2022-NGIoT session_Micha vor dem Berge presentationIoT Week 2022-NGIoT session_Micha vor dem Berge presentation
IoT Week 2022-NGIoT session_Micha vor dem Berge presentation
 
Next Generation IoT Architectures_Hans Salomonsson
Next Generation IoT Architectures_Hans SalomonssonNext Generation IoT Architectures_Hans Salomonsson
Next Generation IoT Architectures_Hans Salomonsson
 
CONASENSE 2022_Jens Hagemeyer presentation
CONASENSE 2022_Jens Hagemeyer presentationCONASENSE 2022_Jens Hagemeyer presentation
CONASENSE 2022_Jens Hagemeyer presentation
 
NGIoT standardisation workshops_Jens Hagemeyer presentation
NGIoT standardisation workshops_Jens Hagemeyer presentationNGIoT standardisation workshops_Jens Hagemeyer presentation
NGIoT standardisation workshops_Jens Hagemeyer presentation
 
IoT Tech Expo 2023_Pedro Trancoso presentation
IoT Tech Expo 2023_Pedro Trancoso presentationIoT Tech Expo 2023_Pedro Trancoso presentation
IoT Tech Expo 2023_Pedro Trancoso presentation
 
HiPEAC-CSW 2022_Kevin Mika presentation
HiPEAC-CSW 2022_Kevin Mika presentationHiPEAC-CSW 2022_Kevin Mika presentation
HiPEAC-CSW 2022_Kevin Mika presentation
 
HiPEAC 2022-DL4IoT workshop_René Griessl presentation
HiPEAC 2022-DL4IoT workshop_René Griessl presentationHiPEAC 2022-DL4IoT workshop_René Griessl presentation
HiPEAC 2022-DL4IoT workshop_René Griessl presentation
 
SS-CPSIoT 2023_Kevin Mika and Piotr Zierhoffer presentation
SS-CPSIoT 2023_Kevin Mika and Piotr Zierhoffer presentationSS-CPSIoT 2023_Kevin Mika and Piotr Zierhoffer presentation
SS-CPSIoT 2023_Kevin Mika and Piotr Zierhoffer presentation
 
HiPEAC2023-DL4IoT Workshop_Jean Hagemeyer presentation
HiPEAC2023-DL4IoT Workshop_Jean Hagemeyer presentationHiPEAC2023-DL4IoT Workshop_Jean Hagemeyer presentation
HiPEAC2023-DL4IoT Workshop_Jean Hagemeyer presentation
 
IoT Week 2021_Jens Hagemeyer presentation
IoT Week 2021_Jens Hagemeyer presentationIoT Week 2021_Jens Hagemeyer presentation
IoT Week 2021_Jens Hagemeyer presentation
 
HiPEAC 2022_Marcelo Pasin presentation
HiPEAC 2022_Marcelo Pasin presentationHiPEAC 2022_Marcelo Pasin presentation
HiPEAC 2022_Marcelo Pasin presentation
 
IoT Tech Expo 2023_Marcelo Pasin presentation
IoT Tech Expo 2023_Marcelo Pasin presentationIoT Tech Expo 2023_Marcelo Pasin presentation
IoT Tech Expo 2023_Marcelo Pasin presentation
 
IoT Tech Expo 2023_Hans-Martin Heyn presentation
IoT Tech Expo 2023_Hans-Martin Heyn presentationIoT Tech Expo 2023_Hans-Martin Heyn presentation
IoT Tech Expo 2023_Hans-Martin Heyn presentation
 
HiPEAC 2022_Marco Tassemeier presentation
HiPEAC 2022_Marco Tassemeier presentationHiPEAC 2022_Marco Tassemeier presentation
HiPEAC 2022_Marco Tassemeier presentation
 
HiPEAC Computing Systems Week 2022_Mario Porrmann presentation
HiPEAC Computing Systems Week 2022_Mario Porrmann presentationHiPEAC Computing Systems Week 2022_Mario Porrmann presentation
HiPEAC Computing Systems Week 2022_Mario Porrmann presentation
 
NGIoT Sustainability Workshop 2023_ Hans-Martin Heyn presentation
NGIoT Sustainability Workshop 2023_ Hans-Martin Heyn presentationNGIoT Sustainability Workshop 2023_ Hans-Martin Heyn presentation
NGIoT Sustainability Workshop 2023_ Hans-Martin Heyn presentation
 
EU-IoT Training Workshops Series: AIoT and Edge Machine Learning 2021_Jens Ha...
EU-IoT Training Workshops Series: AIoT and Edge Machine Learning 2021_Jens Ha...EU-IoT Training Workshops Series: AIoT and Edge Machine Learning 2021_Jens Ha...
EU-IoT Training Workshops Series: AIoT and Edge Machine Learning 2021_Jens Ha...
 

Recently uploaded

Animal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptxAnimal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptxUmerFayaz5
 
Analytical Profile of Coleus Forskohlii | Forskolin .pdf
Analytical Profile of Coleus Forskohlii | Forskolin .pdfAnalytical Profile of Coleus Forskohlii | Forskolin .pdf
Analytical Profile of Coleus Forskohlii | Forskolin .pdfSwapnil Therkar
 
BIOETHICS IN RECOMBINANT DNA TECHNOLOGY.
BIOETHICS IN RECOMBINANT DNA TECHNOLOGY.BIOETHICS IN RECOMBINANT DNA TECHNOLOGY.
BIOETHICS IN RECOMBINANT DNA TECHNOLOGY.PraveenaKalaiselvan1
 
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptx
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptxSOLUBLE PATTERN RECOGNITION RECEPTORS.pptx
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptxkessiyaTpeter
 
Analytical Profile of Coleus Forskohlii | Forskolin .pptx
Analytical Profile of Coleus Forskohlii | Forskolin .pptxAnalytical Profile of Coleus Forskohlii | Forskolin .pptx
Analytical Profile of Coleus Forskohlii | Forskolin .pptxSwapnil Therkar
 
Biopesticide (2).pptx .This slides helps to know the different types of biop...
Biopesticide (2).pptx  .This slides helps to know the different types of biop...Biopesticide (2).pptx  .This slides helps to know the different types of biop...
Biopesticide (2).pptx .This slides helps to know the different types of biop...RohitNehra6
 
Recombination DNA Technology (Nucleic Acid Hybridization )
Recombination DNA Technology (Nucleic Acid Hybridization )Recombination DNA Technology (Nucleic Acid Hybridization )
Recombination DNA Technology (Nucleic Acid Hybridization )aarthirajkumar25
 
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...anilsa9823
 
Work, Energy and Power for class 10 ICSE Physics
Work, Energy and Power for class 10 ICSE PhysicsWork, Energy and Power for class 10 ICSE Physics
Work, Energy and Power for class 10 ICSE Physicsvishikhakeshava1
 
Is RISC-V ready for HPC workload? Maybe?
Is RISC-V ready for HPC workload? Maybe?Is RISC-V ready for HPC workload? Maybe?
Is RISC-V ready for HPC workload? Maybe?Patrick Diehl
 
Genomic DNA And Complementary DNA Libraries construction.
Genomic DNA And Complementary DNA Libraries construction.Genomic DNA And Complementary DNA Libraries construction.
Genomic DNA And Complementary DNA Libraries construction.k64182334
 
Disentangling the origin of chemical differences using GHOST
Disentangling the origin of chemical differences using GHOSTDisentangling the origin of chemical differences using GHOST
Disentangling the origin of chemical differences using GHOSTSérgio Sacani
 
Hubble Asteroid Hunter III. Physical properties of newly found asteroids
Hubble Asteroid Hunter III. Physical properties of newly found asteroidsHubble Asteroid Hunter III. Physical properties of newly found asteroids
Hubble Asteroid Hunter III. Physical properties of newly found asteroidsSérgio Sacani
 
Luciferase in rDNA technology (biotechnology).pptx
Luciferase in rDNA technology (biotechnology).pptxLuciferase in rDNA technology (biotechnology).pptx
Luciferase in rDNA technology (biotechnology).pptxAleenaTreesaSaji
 
Physiochemical properties of nanomaterials and its nanotoxicity.pptx
Physiochemical properties of nanomaterials and its nanotoxicity.pptxPhysiochemical properties of nanomaterials and its nanotoxicity.pptx
Physiochemical properties of nanomaterials and its nanotoxicity.pptxAArockiyaNisha
 
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...Sérgio Sacani
 
GFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptxGFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptxAleenaTreesaSaji
 
Scheme-of-Work-Science-Stage-4 cambridge science.docx
Scheme-of-Work-Science-Stage-4 cambridge science.docxScheme-of-Work-Science-Stage-4 cambridge science.docx
Scheme-of-Work-Science-Stage-4 cambridge science.docxyaramohamed343013
 

Recently uploaded (20)

Animal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptxAnimal Communication- Auditory and Visual.pptx
Animal Communication- Auditory and Visual.pptx
 
Analytical Profile of Coleus Forskohlii | Forskolin .pdf
Analytical Profile of Coleus Forskohlii | Forskolin .pdfAnalytical Profile of Coleus Forskohlii | Forskolin .pdf
Analytical Profile of Coleus Forskohlii | Forskolin .pdf
 
Engler and Prantl system of classification in plant taxonomy
Engler and Prantl system of classification in plant taxonomyEngler and Prantl system of classification in plant taxonomy
Engler and Prantl system of classification in plant taxonomy
 
BIOETHICS IN RECOMBINANT DNA TECHNOLOGY.
BIOETHICS IN RECOMBINANT DNA TECHNOLOGY.BIOETHICS IN RECOMBINANT DNA TECHNOLOGY.
BIOETHICS IN RECOMBINANT DNA TECHNOLOGY.
 
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptx
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptxSOLUBLE PATTERN RECOGNITION RECEPTORS.pptx
SOLUBLE PATTERN RECOGNITION RECEPTORS.pptx
 
Analytical Profile of Coleus Forskohlii | Forskolin .pptx
Analytical Profile of Coleus Forskohlii | Forskolin .pptxAnalytical Profile of Coleus Forskohlii | Forskolin .pptx
Analytical Profile of Coleus Forskohlii | Forskolin .pptx
 
Biopesticide (2).pptx .This slides helps to know the different types of biop...
Biopesticide (2).pptx  .This slides helps to know the different types of biop...Biopesticide (2).pptx  .This slides helps to know the different types of biop...
Biopesticide (2).pptx .This slides helps to know the different types of biop...
 
The Philosophy of Science
The Philosophy of ScienceThe Philosophy of Science
The Philosophy of Science
 
Recombination DNA Technology (Nucleic Acid Hybridization )
Recombination DNA Technology (Nucleic Acid Hybridization )Recombination DNA Technology (Nucleic Acid Hybridization )
Recombination DNA Technology (Nucleic Acid Hybridization )
 
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...
Lucknow 💋 Russian Call Girls Lucknow Finest Escorts Service 8923113531 Availa...
 
Work, Energy and Power for class 10 ICSE Physics
Work, Energy and Power for class 10 ICSE PhysicsWork, Energy and Power for class 10 ICSE Physics
Work, Energy and Power for class 10 ICSE Physics
 
Is RISC-V ready for HPC workload? Maybe?
Is RISC-V ready for HPC workload? Maybe?Is RISC-V ready for HPC workload? Maybe?
Is RISC-V ready for HPC workload? Maybe?
 
Genomic DNA And Complementary DNA Libraries construction.
Genomic DNA And Complementary DNA Libraries construction.Genomic DNA And Complementary DNA Libraries construction.
Genomic DNA And Complementary DNA Libraries construction.
 
Disentangling the origin of chemical differences using GHOST
Disentangling the origin of chemical differences using GHOSTDisentangling the origin of chemical differences using GHOST
Disentangling the origin of chemical differences using GHOST
 
Hubble Asteroid Hunter III. Physical properties of newly found asteroids
Hubble Asteroid Hunter III. Physical properties of newly found asteroidsHubble Asteroid Hunter III. Physical properties of newly found asteroids
Hubble Asteroid Hunter III. Physical properties of newly found asteroids
 
Luciferase in rDNA technology (biotechnology).pptx
Luciferase in rDNA technology (biotechnology).pptxLuciferase in rDNA technology (biotechnology).pptx
Luciferase in rDNA technology (biotechnology).pptx
 
Physiochemical properties of nanomaterials and its nanotoxicity.pptx
Physiochemical properties of nanomaterials and its nanotoxicity.pptxPhysiochemical properties of nanomaterials and its nanotoxicity.pptx
Physiochemical properties of nanomaterials and its nanotoxicity.pptx
 
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
All-domain Anomaly Resolution Office U.S. Department of Defense (U) Case: “Eg...
 
GFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptxGFP in rDNA Technology (Biotechnology).pptx
GFP in rDNA Technology (Biotechnology).pptx
 
Scheme-of-Work-Science-Stage-4 cambridge science.docx
Scheme-of-Work-Science-Stage-4 cambridge science.docxScheme-of-Work-Science-Stage-4 cambridge science.docx
Scheme-of-Work-Science-Stage-4 cambridge science.docx
 

HiPEAC2022_António Casimiro presentation

  • 1. ML Robustness in VEDLIoT António Casimiro University of Lisbon HiPEAC 2022 Budapest, 20 June 2022
  • 2. 2  Local monitoring of input data correctness  Check characteristics of input features (general data)  Build ECDF of training data  Compare with ECDF of input data  Find outliers/drifts in data values (time series data)  Store input data (time series values)  Forecast next expected input value  Compare forecast value with received input value Robustness and safety service  Remote monitoring of model correctness  Replicated execution using same input on trusted model (not for time series data)  Periodically send input/output data to remote node  Run trusted model with received input data  Compare outputs of local and remote models
  • 3. 3  Aware of a context change  Measure statistical distance between  Training data distribution  Real-world data distribution  If huge difference in distributions -> Do not trust ML output Domain monitoring using statistical methods  Calculate empirical cumulative distribution function (ECDF) 1. Ordering all unique observations in data sample 2. Calculating cumulative probability for each as number of observations less than or equal to a given observation divided by the total number of observations:
  • 4. 4  In practice  Take 30 images from training set and 30 images from real-world  Do that for a specific class images based of classifier prediction  For each image take the first pixel in the left corner  Considered one RGB color channel at time  Calculate first ECDF on the values of the 30 pixels in the left corner from training set  The same for real-world image -> second ECDF Statistical distance between data distributions  Apply statistical distance and save the value  To that for all the pixels and for the three color channel  Average distances per color channel and compare the distances with a given threshold  Limitations  Too constrained by the color  Differences in brightness can fool the method  Images should be well aligned for proper comparison
  • 5. 5  Specific models for different environmental conditions  Can they perform better than a single generic one?  Experiment with CNN object detector (Yolo4)  Two different conditions: daytime and night  Driving dataset with 10 classes (pedestrian, rider, car, …) Domain adaptation through split approach  Training three models:  Daytime model, only objects during daytime  Night model, only objects during night  Daytime and night model, both previous ones  Testing models:  On daytime images only  On night images only Daytime + Night Daytime Night Daytime images: 27,967 Night images: 27,967 Daytime + night images: 55,934 Night images: 3,929 Daytime images: 3,929 Training Testing
  • 6. 6  Object detector predicts:  Location of object (coordinates)  Class (e.g., a dog)  Confidence score (a value from 0 to 1)  Confidence score measures the confidence on:  Localization  How likely the box contains an object  How accurate is the box -> IoU  Classification  Precision  «When it guesses how often does it guess correctly?»  Recall  «Has it guessed every time that it should have guessed?» Performance metric  Confidence threshold  Positive detection if confidence score > threshold  Strict threshold -> less recall  Precision-recall (PR) curve  Shows trade-off precision/recall for varying threshold  mean Average precision (mAP)  Summarize such plot for all classes  mAP@0.5 at IoU threshold of 50%
  • 7. 7  Day and night model performs slightly better in both conditions  Trained twice images than the other two  It “saw” more objects during training  Better a single model and train as much data as possible  Then monitor the output Results  Other ways to improve robustness and ensure correctness of output  Adversarial training  Uncertainty quantification  Explainability methods  Statistical distance between data distributions Performance on mAP@0.5 Day test set Night test set Day model 48.59% Night model 47.82% Day and night model 50.58% 49.90%
  • 8. 8 Input monitoring (time series data)  Framework with offline part (model training) and online part (error detection in input data)  Detected errors (due to sensor or communication faults):  Omissions  Outliers  Drifts Self Neighbors Self + Neighbors Generated Models Offline Online
  • 9. 9 Self Neighbors Self + Neighbors Generated Models Offline Training phase  Multiple MLP models are trained to forecast the next data value to be received by a target sensor:  A model using only past data from the target sensor  A model using data from the target sensor and from neighbor sensors (if they provide correlated data)  A model using only data from neighbor sensors (if they provide correlated data)  This approach allows to distinguish real events (with impact on multiple sensors) from outliers (affecting only the target sensor)
  • 10. 10 Online Online error detection  Sensor values go through the monitoring service  Omissions are detected using timers configured for periodic data  Sensor data (from multiple sensors) is stored and properly aligned to feed each model input  Using multiple forecasts (from running the multiple models), outliers can be detected  The service can also replace outliers for quality assurance
  • 12. 12  Monitoring output through explainability  Provide more evidence to assess output correctness  Explainability methods are used to evaluate contribution of each input pixel to the output  Monitoring model correctness  Periodically send images to redundant remote trustworthy model for comparison of results  Safety requirements on the architectural framework  Safety is one of the clusters of concern addressed at several levels of abstration Other ongoing work
  • 13. 13 Conclusion Robustness and safety are important concerns, being addressed from several perspectives • Monitoring methods for input and output data quality • Monitoring methods for checking model correctness • Architecting for safety

Editor's Notes

  1. This showed useful for the quality block as it made possible to correlate events between sensors and distinguish environmental events from sensor errors. However, they are only used when we have more than one sensor.
  2. Intro The integration of safety-critical systems brings a deep reflection on how to guarantee safety and preserve society from an accident. In contrast to typical software, ML control flows are specified by inscrutable weights and trained and tested pointwise using specific cases, which has limited effectiveness at improving and assessing an ML system’s completeness and coverage. They rarely manage all test cases (uncertainty) and are susceptible to small changes in input (adversarial examples), even if they give us an higher confidence score for a prediction (score produced by the model that expresses how confident it is of the prediction correctness). Flowchart --- off-line phase --- Select the most suitable dataset for our goal. It is assumed that it is used a trusted dataset to train, validate and test the model. So it is correctly labelled and the images for classes are balanced. Data augmentation techniques can be used to increase the amount of data. 2) Apply Ranger technique for hardware transient faults propagation prevention (e.g., bit-flip), which consists in applying limits to the ranges of the output values from the activation functions of the neural network. 3) Train the model. 4) Generate adversarial examples to increase the adversarial robustness by retraining the model with these examples. 5) Evaluate the model's performance. If we consider satisfactory, we can use it into the autonomous system. Otherwise it will be necessary to change model or refine the data, and then repeat the training process. 6) Use explainability methods on the model and the test set to evaluate the contribution of each input feature to the output assigning an importance score to each individual feature (each pixel). 7) Use this data to train a second model, with the same architecture as the first, which should give us more information on the output of the main model. Also in this case we evaluate the performance of the model and if it is not satisfactory we can consider adding further data. --- on-line phase --- 8) Here the main model will be monitored during its real-time operation. The data from real world is going to be feed to the system and the model will return predictions with relative confidence score. 9) Each input will be passed to both models and the outputs for this input will be compared. 10.1) If both gave the same result and the estimated uncertainty is below a certain threshold, then we can trust the result of the main model and use this prediction for the subsequent operations of the system. 10.2) If the uncertainty is not higher than the threshold, but the two models give a different result then the result of the second model is kept if the confidence score is high enough, otherwise we must pass the control to the human. 10.3) If the uncertainty is high, the comparison between the two outputs is not taken into account and also in this case the control is given to the human.