Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
FlexiBO: A Decoupled Cost-Aware Multi-Objective Optimization of Deep Neural Networks
Search
Pooyan Jamshidi
February 29, 2024
Science
0
48
FlexiBO: A Decoupled Cost-Aware Multi-Objective Optimization of Deep Neural Networks
AAAI 2024
Pooyan Jamshidi
February 29, 2024
Tweet
Share
More Decks by Pooyan Jamshidi
See All by Pooyan Jamshidi
Reconciling Accuracy, Cost, and Latency of Inference Serving Systems
pjamshidi
0
26
Reconciling High Accuracy, Cost-Efficiency, and Low Latency of Inference Serving Systems
pjamshidi
0
54
Learning from Valerie Issarny: Insights Gained from Program Co-Chairing SEAMS’23
pjamshidi
0
150
Artificial Intelligence and Systems Laboratory (AISys): A Research Overview
pjamshidi
0
380
Experiential Learning by Building Real-World AI Systems
pjamshidi
0
150
Understanding and Explaining the Root Causes of Performance Faults with Causal AI: A Path towards Building Dependable Computer Systems
pjamshidi
0
100
On Debugging the Performance of Configurable Software Systems: Developer Needs and Tailored Tool Support
pjamshidi
0
190
Unicorn: Reasoning about Configurable System Performance through the Lens of Causality
pjamshidi
0
370
Causal AI for Systems
pjamshidi
0
240
Other Decks in Science
See All in Science
ICRA2024 速報
rpc
3
4.4k
Machine Learning for Materials (Lecture 4)
aronwalsh
0
730
The thin line between reconstruction, classification, and hallucination in brain decoding
ykamit
1
430
大規模言語モデルの開発
chokkan
PRO
80
27k
HAS Dark Site Orientation
astronomyhouston
0
5.1k
Pandas 2 vs Polars vs Dask (PyDataGlobal 2023 December)
ianozsvald
0
660
Raccoon Roundworm
uni_of_nomi
0
110
Snowflake上でRを使う: RStudioセットアップとShinyアプリケーションのデプロイ
ktatsuya
0
260
Презентация программы бакалавриата СПбГУ "Искусственный интеллект и наука о данных"
dscs
0
400
Machine Learning for Materials (Lecture 9)
aronwalsh
0
170
様々な侵入者タイプに対応した適切な警備計画の策定 / Patrol route design considering various types of intrudes
konakalab
0
120
文系出身でも「アルゴリズム×数学」はスッキリ理解できた!話
wakamatsu_takumu
0
300
Featured
See All Featured
A Modern Web Designer's Workflow
chriscoyier
689
190k
Reflections from 52 weeks, 52 projects
jeffersonlam
346
19k
From Idea to $5000 a Month in 5 Months
shpigford
377
46k
RailsConf 2023
tenderlove
16
720
Git: the NoSQL Database
bkeepers
PRO
423
64k
Cheating the UX When There Is Nothing More to Optimize - PixelPioneers
stephaniewalter
277
13k
No one is an island. Learnings from fostering a developers community.
thoeni
17
2.8k
Web Components: a chance to create the future
zenorocha
307
41k
Statistics for Hackers
jakevdp
792
220k
Sharpening the Axe: The Primacy of Toolmaking
bcantrill
26
1.6k
Six Lessons from altMBA
skipperchong
24
3.2k
Keith and Marios Guide to Fast Websites
keithpitt
408
22k
Transcript
FlexiBO: A Decoupled Cost-Aware Multi-Objective Optimization of Deep Neural Networks
Shahriar Iqbal, Jianhai Su, Lars Kotthoff, Pooyan Jamshidi
[email protected]
AAAI, 24 February 2024 1
One Size Does Not Fit All 1 1.5 2 2.5
3 3.5 ·104 15 20 25 30 35 40 Energy Consumption (mJ) Prediction Error (%) Xception ← Energy consumption varies 4 × → ← Prediction Error varies 3 × → 2
Heterogeneous Parameters Num of Filters, Filter Size, Learning Rate, Num
of Epochs DN N Design Compiler Hardware Deployment Num of Active CPUs, CPU/ GPU/ EMC Frequency Cloud, IoT, Edge Num of Threads, GPU Threads, Memory Growth 3
Cost-Unaware Methods Waste Resources Coupled Unaware Pareto Optimal Prediction Error
(%) Log Wall Clock Time Energy Consumption (mJ) 3000 6000 9000 12000 15 25 35 45 3.65 3.50 3.35 Decoupled Aware Pareto Optimal Prediction Error (%) Log Wall Clock Time Energy Consumption (mJ) 3000 6000 9000 12000 15 25 35 45 3.65 3.50 3.35 4
Proposed Method ▷ weight expected benefit of evaluation by cost
▷ choose which objective(s) to evaluate ▷ more efficient use of resources – lower cost, more evaluations 5
Results – Computer Vision 0 50 100 150 200 Cumulative
Log WallClock Time 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 Hypervolume Error Xception PAL PESMO ParEGO SMSEGO CA-MOBO PESMO-DEC FLEXIBO-GPLC 10000 15000 20000 25000 Energy Consumption (mJ) 15 20 25 30 35 40 Prediction Error (%) Xception PAL PESMO ParEGO SMSEGO CA-MOBO PESMO-DEC FLEXIBO-GPLC 6
Results – NLP 0 50 100 150 200 Cumulative Log
WallClock Time 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 Hypervolume Error BERT-SQuAD PAL PESMO ParEGO SMSEGO CA-MOBO PESMO-DEC FLEXIBO-GPLC 20000 30000 40000 50000 60000 70000 80000 90000 Energy Consumption (mJ) 20 25 30 35 Prediction Error (%) BERT-SQuAD PAL PESMO ParEGO SMSEGO CA-MOBO PESMO-DEC FLEXIBO-GPLC 7
Results – Speech Recognition 0 50 100 150 200 250
300 Cumulative Log WallClock Time 0.25 0.30 0.35 0.40 0.45 0.50 0.55 Hypervolume Error DeepSpeech PAL PESMO ParEGO SMSEGO CA-MOBO PESMO-DEC FLEXIBO-GPLC 20000 30000 40000 50000 60000 Energy Consumption (mJ) 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0 Prediction Error (%) DeepSpeech PAL PESMO ParEGO SMSEGO CA-MOBO PESMO-DEC FLEXIBO-GPLC 8
Results – Evaluations 0 20 40 60 80 100 120
140 160 180 200 PAL 0 20 40 60 80 100 120 140 160 180 200 PESMO-DEC 2 4 6 8 0 20 40 60 80 100 120 140 160 180 200 Iteration CA-MOBO 0 20 40 60 80 100 120 140 160 180 200 Iteration FlexiBO 2 4 6 8 9
FlexiBO: A Decoupled Cost-Aware Multi-Objective Optimization of Deep Neural Networks
▷ cost-aware acquisition function decreases cost and improves results ▷ code available at https://github.com/softsys4ai/FlexiBO 0 50 100 150 200 250 300 Cumulative Log WallClock Time 0.25 0.30 0.35 0.40 0.45 0.50 0.55 Hypervolume Error DeepSpeech PAL PESMO ParEGO SMSEGO CA-MOBO PESMO-DEC FLEXIBO-GPLC 20000 30000 40000 50000 60000 Energy Consumption (mJ) 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0 Prediction Error (%) DeepSpeech PAL PESMO ParEGO SMSEGO CA-MOBO PESMO-DEC FLEXIBO-GPLC 10