SlideShare uma empresa Scribd logo
1 de 18
Lightweight Deep Learning
on Edge Devices
Veljko Pejović (veljko.pejovic@fri.uni-lj.si)
Faculty of Computer and Information Science
University of Ljubljana, Slovenia
Computer Science Department,
Lancaster University, UK
AI Should Live on the Edge
Privacy and availability
“4 in 10 consumers opt not to use the [AI-powered
voice assistant] services because they are worried
about their data”
The Voice Consumer Index (VCI)
Vixen Labs, 2021
“AI requires a high-bandwidth, low-latency network.
It is important to ensure the service wrap and
technology stack are consistent for all regions”
What are the infrastructure requirements for artificial intelligence?
Terry Storrar, Leaseweb, 2021
AI Struggles on the Edge
Latency, memory, energy
• Limited resources vs
increasing model requirements
Canziani, A., Paszke, A., & Culurciello, E. (2016). An analysis
of deep neural network models for practical applications.
arXiv preprint arXiv:1605.07678.
• Heterogeneous devices and
latency/energy burden
Wang, H., Kim, B., Xie, J., & Han, Z.
How is energy consumed in smartphone deep learning apps?
Executing locally vs. remotely. In IEEE GLOBECOM 2019
Next Generation Hardware Won’t Help
Mobiles will lag
• Breakdown of Dennard scaling
• Packing more transistors in the
same area will dissipate more power
• Multicore needs space
• More energy for computation and cooling
[Hennessy & Patterson, Turing Award Lecture 2019]
Approximate Mobile Computing
We don’t need all the computing power and energy all the time
AMC
Opportunities for AMC
• Computed result quality exceeds the limits of human perception or attention
• Computed result quality exceeds a user’s interest/need
• Preserving resources is more important than high result quality
• Inputs and/or the computation are inherently noisy
• Inputs are inherently “easy” to process
Bringing AMC to Masses
Programming support for context-aware approximation
• All developers should be able to approximate
• Mobile developers are not data scientists
• Approximation should be dynamic
Mobiprox
Supporting approximate deep learning on mobiles
• Implement support for approximate
tensor operations on Android
M. Fabjančič, O. Machidon, H. Sharif, Y. Zhao, S. Misailović, V. Pejović
Mobiprox: Supporting Dynamic Approximate Computing on Mobiles
arXiv:2303.11291 (2023)
Mobiprox
Supporting approximate deep learning on mobiles
QoS loss
Speedup
• Implement support for approximate
tensor operations on Android
• Uncover the Pareto-front of
configurations (layer-wise
approximations), that give the optimal
speedup — inference accuracy trade-
off
• Devise dynamic adaptation
algorithms for navigating the Pareto
front
Mobiprox
Supporting approximate deep learning on mobiles
• Approximations:
• Filter sampling, perforated convolutions, quantization
• Implementation: expanded CLBlast lib
• Tuning:
• On a GPU-enabled cluster * **
• On an Android device
Row perforation and column perforation
Filter sampling
* Sharif et al., ApproxTuner: A Compiler and Runtime System for Adaptive Approximations. PPoPP, 2021
** Sharif et al. ApproxHPVM: a portable compiler IR for accuracy-aware optimizations. OOPSLA, 2019
Profiler for
Android
ApproxTuner
Mobiprox
Compiler pipeline
PyTorch
- Definition
- Training
Tuning binary
(CUDA)
Inference binary
(OpenCL)
ART
JNI
Preparation
LLVM
hpvm-tensor-
rt-android
Java/Kotlin
App
Dynamic Approximation Adaptation
Context-aware, need-driven, business-oriented adaptation
• Arbitrary adaptation strategies can be implemented
• “More accurate human activity recognition model when a user is exercising”
• “Higher approximation level when battery falls under 15%”
• Our pick: “Minimize energy usage without sacrificing the inference accuracy”
Dynamic Approximation Adaptation
Driven by SoftMax confidence thresholds
SoftMax
confidence
t
Evaluation
Human activity recognition
• 21 volunteers, on-body UDOO boards,
six prescribed activities
• Slight accuracy drop
from 65% to 63% accuracy (-2%)
• Significant energy savings
from 245mAh to 209mAh (-15%)
• Certain classes are more robust
to approximation than others
Average accuracy vs. average energy consumption for all users
non-approximated network vs confidence-based adaptation
Evaluation
Spoken keyword recognition
• HONK model built on Google SC
• Mix 160 unheard utterances from
Google SC with noise levels from
realistic environments
• Confidence-based adaptation
• 15% less energy, 0% accuracy loss
Acknowledgements
The Team Resources
• Octavian Machidon
• Alina Machidon
• Davor Sluga
• Matevž Fabjančič
• Timotej Knez
• Janez Božič
• Tine Fajfar
• Jani Asprov
“Bringing Resource Efficiency to Smartphones with Approximate
Computing”
(ARRS project No.: N2-0136)
“Context-Aware On-Device Approximate Computing”
(ARRS project No.: J2-3047)
“Computer Structures and Systems”
(ARRS core funding No. P2-0098.
M. Fabjancic et al. Mobiprox: Supporting Dynamic Approximate Computing on
Mobiles, arXiv:2303.11291, 2023
A. Machidon and V. Pejovic, Enabling Resource-Efficient Edge Intelligence
with Compressive Sensing-Based Deep Learning, ACM Computing Frontiers,
May 2022
A. Machidon and V. Pejovic, Deep Learning Techniques for Compressive
Sensing-Based Reconstruction and Inference - A Ubiquitous Systems
Perspective, Artificial Intelligence Review, 2022
T. Knez, O. Machidon, and V. Pejovic, Self-Adaptive Approximate Mobile Deep
Learning, Electronics (2021)
V. Pejovic, Towards Approximate Mobile Computing, ACM GetMobile
Magazine, Vol 22(5), December, 2018.
Thank you!
Veljko Pejović (veljko.pejovic@fri.uni-lj.si)
University of Ljubljana, Slovenia
Lancaster University, UK
Code available at https://gitlab.fri.uni-lj.si/lrk
[DSC Adria 23] Veljko Pejovic Lightweight Deep Learning on Edge Devices.pptx

Mais conteúdo relacionado

Semelhante a [DSC Adria 23] Veljko Pejovic Lightweight Deep Learning on Edge Devices.pptx

Arpan pal ncccs
Arpan pal ncccsArpan pal ncccs
Arpan pal ncccs
Arpan Pal
 
Reducing energy consumption of computing
Reducing energy consumption of computing Reducing energy consumption of computing
Reducing energy consumption of computing
NGUYEN VAN LUONG
 
00 logistics zhu
00 logistics zhu00 logistics zhu
00 logistics zhu
c09271
 

Semelhante a [DSC Adria 23] Veljko Pejovic Lightweight Deep Learning on Edge Devices.pptx (20)

Lecture_IIITD.pptx
Lecture_IIITD.pptxLecture_IIITD.pptx
Lecture_IIITD.pptx
 
Novi sad ai event 1-2018
Novi sad ai event 1-2018Novi sad ai event 1-2018
Novi sad ai event 1-2018
 
Keynote on Mobile Grid and Cloud Computing
Keynote on Mobile Grid and Cloud ComputingKeynote on Mobile Grid and Cloud Computing
Keynote on Mobile Grid and Cloud Computing
 
System Support for Internet of Things
System Support for Internet of ThingsSystem Support for Internet of Things
System Support for Internet of Things
 
Arpan pal ncccs
Arpan pal ncccsArpan pal ncccs
Arpan pal ncccs
 
Edge computing and its role in architecting IoT
Edge computing and its role in architecting IoTEdge computing and its role in architecting IoT
Edge computing and its role in architecting IoT
 
Machine Learning for Multimedia and Edge Information Processing.pptx
Machine Learning for Multimedia and Edge Information Processing.pptxMachine Learning for Multimedia and Edge Information Processing.pptx
Machine Learning for Multimedia and Edge Information Processing.pptx
 
Reducing energy consumption of computing
Reducing energy consumption of computing Reducing energy consumption of computing
Reducing energy consumption of computing
 
00 logistics zhu
00 logistics zhu00 logistics zhu
00 logistics zhu
 
Smart Connectivity
Smart ConnectivitySmart Connectivity
Smart Connectivity
 
Location Data - Finding the needle in the haystack
Location Data - Finding the needle in the haystackLocation Data - Finding the needle in the haystack
Location Data - Finding the needle in the haystack
 
Tutorial ESWC2011 Building Semantic Sensor Web - 01 - Introduction
Tutorial ESWC2011 Building Semantic Sensor Web - 01 - IntroductionTutorial ESWC2011 Building Semantic Sensor Web - 01 - Introduction
Tutorial ESWC2011 Building Semantic Sensor Web - 01 - Introduction
 
“Accelerating Newer ML Models Using the Qualcomm AI Stack,” a Presentation fr...
“Accelerating Newer ML Models Using the Qualcomm AI Stack,” a Presentation fr...“Accelerating Newer ML Models Using the Qualcomm AI Stack,” a Presentation fr...
“Accelerating Newer ML Models Using the Qualcomm AI Stack,” a Presentation fr...
 
Zühlke Meetup - Mai 2017
Zühlke Meetup - Mai 2017Zühlke Meetup - Mai 2017
Zühlke Meetup - Mai 2017
 
Machine Learning -Based Security Authentication for Wireless Multimedia Network
Machine Learning -Based Security Authentication for Wireless Multimedia NetworkMachine Learning -Based Security Authentication for Wireless Multimedia Network
Machine Learning -Based Security Authentication for Wireless Multimedia Network
 
What is Your Edge From the Cloud to the Edge, Extending Your Reach
What is Your Edge From the Cloud to the Edge, Extending Your ReachWhat is Your Edge From the Cloud to the Edge, Extending Your Reach
What is Your Edge From the Cloud to the Edge, Extending Your Reach
 
Big Crypto for Little Things
Big Crypto for Little ThingsBig Crypto for Little Things
Big Crypto for Little Things
 
The International Journal of Managing Value and Supply Chains ( IJMVSC )
The International Journal of Managing Value and Supply Chains ( IJMVSC ) The International Journal of Managing Value and Supply Chains ( IJMVSC )
The International Journal of Managing Value and Supply Chains ( IJMVSC )
 
International Journal of Ad hoc, Sensor & Ubiquitous Computing (IJASUC)
International Journal of Ad hoc, Sensor & Ubiquitous Computing (IJASUC)International Journal of Ad hoc, Sensor & Ubiquitous Computing (IJASUC)
International Journal of Ad hoc, Sensor & Ubiquitous Computing (IJASUC)
 
Reconfigurable Embedded Systems Applications for Versatile Biomedical Measure...
Reconfigurable Embedded Systems Applications for Versatile Biomedical Measure...Reconfigurable Embedded Systems Applications for Versatile Biomedical Measure...
Reconfigurable Embedded Systems Applications for Versatile Biomedical Measure...
 

Mais de DataScienceConferenc1

[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
DataScienceConferenc1
 
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
DataScienceConferenc1
 
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
DataScienceConferenc1
 

Mais de DataScienceConferenc1 (20)

[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
[DSC MENA 24] Mostafa_Essa_-_Ai_and_cloud.pdf
 
[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...
[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...
[DSC MENA 24] Yasser_El_Bendary - How NLP & LLMs model can excel in comprehen...
 
[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf
[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf
[DSC MENA 24] Medhat_Kandil - Empowering Egypt's AI & Biotechnology Scenes.pdf
 
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
[DSC MENA 24] Youssef_Kamal - Data governance and quality.pdf
 
[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf
[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf
[DSC MENA 24] Abdelrahman_Ghallab_-_Data_Product_mgmt.pdf
 
[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx
[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx
[DSC MENA 24] Asmaa_Eltaher_-_Innovation_Beyond_Brainstorming.pptx
 
[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf
[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf
[DSC MENA 24] Muhammad_Ezzat_-_Sustianable_Growth_Empowerment.pdf
 
[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...
[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...
[DSC MENA 24] Basma_Rady_-_Building_a_Data_Driven_Culture_in_Your_Organizatio...
 
[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf
[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf
[DSC MENA 24] Ahmed_Muselhy_-_Unveiling-the-Secrets-of-AI-in-Hiring.pdf
 
[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...
[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...
[DSC MENA 24] Ziad_Diab_-_Data-Driven_Disruption_-_The_Role_of_Data_Strategy_...
 
[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...
[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...
[DSC MENA 24] Mohammad_Essam_- Leveraging Scene Graphs for Generative AI and ...
 
[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf
[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf
[DSC MENA 24] Ahmed_Fahmy - Navigating the Future.pdf
 
[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx
[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx
[DSC MENA 24] Hany_Saad_Gheit_-_Azure_OpenAI_service.pptx
 
[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...
[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...
[DSC MENA 24] Nezar_El_Kady_-_From_Turing_to_Transformers__Navigating_the_AI_...
 
[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx
[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx
[DSC MENA 24] Amira_Abdelaziz_-_AI_in_Financial_Services.pptx
 
[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...
[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...
[DSC MENA 24] Omar_Ossama - My Journey from the Field of Oil & Gas, to the Ex...
 
[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...
[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...
[DSC MENA 24] Ramy_Agieb_-_Advancements_in_Artificial_Intelligence_for_Cybers...
 
[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx
[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx
[DSC MENA 24] Sohaila_Diab_-_Lets_Talk_Gen_AI_Presentation.pptx
 
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx
 
[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf
[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf
[DSC MENA 24] Abdelrahman_Sleem_-_AI_For_Marketing_DSC.pdf
 

Último

Top profile Call Girls In Latur [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Latur [ 7014168258 ] Call Me For Genuine Models We ...Top profile Call Girls In Latur [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Latur [ 7014168258 ] Call Me For Genuine Models We ...
gajnagarg
 
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
ZurliaSoop
 
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
gajnagarg
 
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
gajnagarg
 
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
vexqp
 
Reconciling Conflicting Data Curation Actions: Transparency Through Argument...
Reconciling Conflicting Data Curation Actions:  Transparency Through Argument...Reconciling Conflicting Data Curation Actions:  Transparency Through Argument...
Reconciling Conflicting Data Curation Actions: Transparency Through Argument...
Bertram Ludäscher
 
Lecture_2_Deep_Learning_Overview-newone1
Lecture_2_Deep_Learning_Overview-newone1Lecture_2_Deep_Learning_Overview-newone1
Lecture_2_Deep_Learning_Overview-newone1
ranjankumarbehera14
 
Jual obat aborsi Bandung ( 085657271886 ) Cytote pil telat bulan penggugur ka...
Jual obat aborsi Bandung ( 085657271886 ) Cytote pil telat bulan penggugur ka...Jual obat aborsi Bandung ( 085657271886 ) Cytote pil telat bulan penggugur ka...
Jual obat aborsi Bandung ( 085657271886 ) Cytote pil telat bulan penggugur ka...
Klinik kandungan
 
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
nirzagarg
 
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
nirzagarg
 
Top profile Call Girls In Vadodara [ 7014168258 ] Call Me For Genuine Models ...
Top profile Call Girls In Vadodara [ 7014168258 ] Call Me For Genuine Models ...Top profile Call Girls In Vadodara [ 7014168258 ] Call Me For Genuine Models ...
Top profile Call Girls In Vadodara [ 7014168258 ] Call Me For Genuine Models ...
gajnagarg
 

Último (20)

Aspirational Block Program Block Syaldey District - Almora
Aspirational Block Program Block Syaldey District - AlmoraAspirational Block Program Block Syaldey District - Almora
Aspirational Block Program Block Syaldey District - Almora
 
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
 
Kings of Saudi Arabia, information about them
Kings of Saudi Arabia, information about themKings of Saudi Arabia, information about them
Kings of Saudi Arabia, information about them
 
Top profile Call Girls In Latur [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Latur [ 7014168258 ] Call Me For Genuine Models We ...Top profile Call Girls In Latur [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Latur [ 7014168258 ] Call Me For Genuine Models We ...
 
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Surabaya ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
 
Ranking and Scoring Exercises for Research
Ranking and Scoring Exercises for ResearchRanking and Scoring Exercises for Research
Ranking and Scoring Exercises for Research
 
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
 
7. Epi of Chronic respiratory diseases.ppt
7. Epi of Chronic respiratory diseases.ppt7. Epi of Chronic respiratory diseases.ppt
7. Epi of Chronic respiratory diseases.ppt
 
Vadodara 💋 Call Girl 7737669865 Call Girls in Vadodara Escort service book now
Vadodara 💋 Call Girl 7737669865 Call Girls in Vadodara Escort service book nowVadodara 💋 Call Girl 7737669865 Call Girls in Vadodara Escort service book now
Vadodara 💋 Call Girl 7737669865 Call Girls in Vadodara Escort service book now
 
Digital Transformation Playbook by Graham Ware
Digital Transformation Playbook by Graham WareDigital Transformation Playbook by Graham Ware
Digital Transformation Playbook by Graham Ware
 
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
Top profile Call Girls In Chandrapur [ 7014168258 ] Call Me For Genuine Model...
 
5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed
5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed
5CL-ADBA,5cladba, Chinese supplier, safety is guaranteed
 
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
 
Reconciling Conflicting Data Curation Actions: Transparency Through Argument...
Reconciling Conflicting Data Curation Actions:  Transparency Through Argument...Reconciling Conflicting Data Curation Actions:  Transparency Through Argument...
Reconciling Conflicting Data Curation Actions: Transparency Through Argument...
 
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24  Building Real-Time Pipelines With FLaNKDATA SUMMIT 24  Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
 
Lecture_2_Deep_Learning_Overview-newone1
Lecture_2_Deep_Learning_Overview-newone1Lecture_2_Deep_Learning_Overview-newone1
Lecture_2_Deep_Learning_Overview-newone1
 
Jual obat aborsi Bandung ( 085657271886 ) Cytote pil telat bulan penggugur ka...
Jual obat aborsi Bandung ( 085657271886 ) Cytote pil telat bulan penggugur ka...Jual obat aborsi Bandung ( 085657271886 ) Cytote pil telat bulan penggugur ka...
Jual obat aborsi Bandung ( 085657271886 ) Cytote pil telat bulan penggugur ka...
 
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
 
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
 
Top profile Call Girls In Vadodara [ 7014168258 ] Call Me For Genuine Models ...
Top profile Call Girls In Vadodara [ 7014168258 ] Call Me For Genuine Models ...Top profile Call Girls In Vadodara [ 7014168258 ] Call Me For Genuine Models ...
Top profile Call Girls In Vadodara [ 7014168258 ] Call Me For Genuine Models ...
 

[DSC Adria 23] Veljko Pejovic Lightweight Deep Learning on Edge Devices.pptx

  • 1. Lightweight Deep Learning on Edge Devices Veljko Pejović (veljko.pejovic@fri.uni-lj.si) Faculty of Computer and Information Science University of Ljubljana, Slovenia Computer Science Department, Lancaster University, UK
  • 2. AI Should Live on the Edge Privacy and availability “4 in 10 consumers opt not to use the [AI-powered voice assistant] services because they are worried about their data” The Voice Consumer Index (VCI) Vixen Labs, 2021 “AI requires a high-bandwidth, low-latency network. It is important to ensure the service wrap and technology stack are consistent for all regions” What are the infrastructure requirements for artificial intelligence? Terry Storrar, Leaseweb, 2021
  • 3. AI Struggles on the Edge Latency, memory, energy • Limited resources vs increasing model requirements Canziani, A., Paszke, A., & Culurciello, E. (2016). An analysis of deep neural network models for practical applications. arXiv preprint arXiv:1605.07678. • Heterogeneous devices and latency/energy burden Wang, H., Kim, B., Xie, J., & Han, Z. How is energy consumed in smartphone deep learning apps? Executing locally vs. remotely. In IEEE GLOBECOM 2019
  • 4. Next Generation Hardware Won’t Help Mobiles will lag • Breakdown of Dennard scaling • Packing more transistors in the same area will dissipate more power • Multicore needs space • More energy for computation and cooling [Hennessy & Patterson, Turing Award Lecture 2019]
  • 5. Approximate Mobile Computing We don’t need all the computing power and energy all the time AMC
  • 6. Opportunities for AMC • Computed result quality exceeds the limits of human perception or attention • Computed result quality exceeds a user’s interest/need • Preserving resources is more important than high result quality • Inputs and/or the computation are inherently noisy • Inputs are inherently “easy” to process
  • 7. Bringing AMC to Masses Programming support for context-aware approximation • All developers should be able to approximate • Mobile developers are not data scientists • Approximation should be dynamic
  • 8. Mobiprox Supporting approximate deep learning on mobiles • Implement support for approximate tensor operations on Android M. Fabjančič, O. Machidon, H. Sharif, Y. Zhao, S. Misailović, V. Pejović Mobiprox: Supporting Dynamic Approximate Computing on Mobiles arXiv:2303.11291 (2023)
  • 9. Mobiprox Supporting approximate deep learning on mobiles QoS loss Speedup • Implement support for approximate tensor operations on Android • Uncover the Pareto-front of configurations (layer-wise approximations), that give the optimal speedup — inference accuracy trade- off • Devise dynamic adaptation algorithms for navigating the Pareto front
  • 10. Mobiprox Supporting approximate deep learning on mobiles • Approximations: • Filter sampling, perforated convolutions, quantization • Implementation: expanded CLBlast lib • Tuning: • On a GPU-enabled cluster * ** • On an Android device Row perforation and column perforation Filter sampling * Sharif et al., ApproxTuner: A Compiler and Runtime System for Adaptive Approximations. PPoPP, 2021 ** Sharif et al. ApproxHPVM: a portable compiler IR for accuracy-aware optimizations. OOPSLA, 2019
  • 11. Profiler for Android ApproxTuner Mobiprox Compiler pipeline PyTorch - Definition - Training Tuning binary (CUDA) Inference binary (OpenCL) ART JNI Preparation LLVM hpvm-tensor- rt-android Java/Kotlin App
  • 12. Dynamic Approximation Adaptation Context-aware, need-driven, business-oriented adaptation • Arbitrary adaptation strategies can be implemented • “More accurate human activity recognition model when a user is exercising” • “Higher approximation level when battery falls under 15%” • Our pick: “Minimize energy usage without sacrificing the inference accuracy”
  • 13. Dynamic Approximation Adaptation Driven by SoftMax confidence thresholds SoftMax confidence t
  • 14. Evaluation Human activity recognition • 21 volunteers, on-body UDOO boards, six prescribed activities • Slight accuracy drop from 65% to 63% accuracy (-2%) • Significant energy savings from 245mAh to 209mAh (-15%) • Certain classes are more robust to approximation than others Average accuracy vs. average energy consumption for all users non-approximated network vs confidence-based adaptation
  • 15. Evaluation Spoken keyword recognition • HONK model built on Google SC • Mix 160 unheard utterances from Google SC with noise levels from realistic environments • Confidence-based adaptation • 15% less energy, 0% accuracy loss
  • 16. Acknowledgements The Team Resources • Octavian Machidon • Alina Machidon • Davor Sluga • Matevž Fabjančič • Timotej Knez • Janez Božič • Tine Fajfar • Jani Asprov “Bringing Resource Efficiency to Smartphones with Approximate Computing” (ARRS project No.: N2-0136) “Context-Aware On-Device Approximate Computing” (ARRS project No.: J2-3047) “Computer Structures and Systems” (ARRS core funding No. P2-0098. M. Fabjancic et al. Mobiprox: Supporting Dynamic Approximate Computing on Mobiles, arXiv:2303.11291, 2023 A. Machidon and V. Pejovic, Enabling Resource-Efficient Edge Intelligence with Compressive Sensing-Based Deep Learning, ACM Computing Frontiers, May 2022 A. Machidon and V. Pejovic, Deep Learning Techniques for Compressive Sensing-Based Reconstruction and Inference - A Ubiquitous Systems Perspective, Artificial Intelligence Review, 2022 T. Knez, O. Machidon, and V. Pejovic, Self-Adaptive Approximate Mobile Deep Learning, Electronics (2021) V. Pejovic, Towards Approximate Mobile Computing, ACM GetMobile Magazine, Vol 22(5), December, 2018.
  • 17. Thank you! Veljko Pejović (veljko.pejovic@fri.uni-lj.si) University of Ljubljana, Slovenia Lancaster University, UK Code available at https://gitlab.fri.uni-lj.si/lrk