Non-Profiled Deep Learning-based Side-Channel attacks with - - PDF document

non profiled deep learning based side channel attacks
SMART_READER_LITE
LIVE PREVIEW

Non-Profiled Deep Learning-based Side-Channel attacks with - - PDF document

Non-Profiled Deep Learning-based Side-Channel attacks with Sensitivity Analysis Benjamin Timon August 28, 2019 Python notebook presentation for CHES 2019 1 Introduction & Motivation Profiled vs Non-Profiled attacks Machine Learning trend


slide-1
SLIDE 1

Non-Profiled Deep Learning-based Side-Channel attacks with Sensitivity Analysis

Benjamin Timon August 28, 2019 Python notebook presentation for CHES 2019

1

slide-2
SLIDE 2

Introduction & Motivation

Profiled vs Non-Profiled attacks Machine Learning trend

2

slide-3
SLIDE 3

This research

3

slide-4
SLIDE 4

Differential Deep Learning Analysis (DDLA)

Correlation Power Analysis Follow similar strategy for DDLA

4

slide-5
SLIDE 5

Demonstration: attack with accuracy and loss

Generate simulation traces

■♥ ❬✶❪✿ ❢r♦♠ ❞❡♠♦ ✐♠♣♦rt ❣❡♥❴s✐♠✉❴❞❛t❛ ★ ●❡♥❡r❛t❡ s✐♠✉❧❛t✐♦♥ tr❛❝❡s ❢♦r ❞❡♠♦♥str❛t✐♦♥ ❞❛t❛❴tr❛✐♥✐♥❣ ❂ ❣❡♥❴s✐♠✉❴❞❛t❛✭✮

Network

5

slide-6
SLIDE 6

Observe loss and accuracy during training

■♥ ❬✷❪✿ ❢r♦♠ ❞❡♠♦ ✐♠♣♦rt ❞❡♠♦❴❞❞❧❛❴❛❝❝❴❧♦ss ❞❞❧❛ ❂ ❞❡♠♦❴❞❞❧❛❴❛❝❝❴❧♦ss✭❞❛t❛❴tr❛✐♥✐♥❣✮ ❞❞❧❛✳r✉♥✭♥❴❡♣♦❝❤s❂✸✵✮ ❞❞❧❛✳❢✐❣ ❖✉t❬✷❪✿

Sensitivity Analysis

6

slide-7
SLIDE 7

Demonstration: Observe first layer gradient

■♥ ❬✸❪✿ ❢r♦♠ ♣❧♦ts ✐♠♣♦rt ♣❧♦t❴✇❡✐❣❤ts❴❛♥❞❴❣r❛❞❴✸❞ ♣❧♦t❴✇❡✐❣❤ts❴❛♥❞❴❣r❛❞❴✸❞✭❞❛t❛❴tr❛✐♥✐♥❣✮ ❖✉t❬✸❪✿

Observe first layer gradient during training

■♥ ❬✹❪✿ ❢r♦♠ ❞❡♠♦ ✐♠♣♦rt ❞❡♠♦❴❞❞❧❛❴❣r❛❞✐❡♥t ❞❞❧❛ ❂ ❞❡♠♦❴❞❞❧❛❴❣r❛❞✐❡♥t✭❞❛t❛❴tr❛✐♥✐♥❣✮ ❞❞❧❛✳r✉♥✭✶✺✮ ❞❞❧❛✳❢✐❣ ❖✉t❬✹❪✿ 7

slide-8
SLIDE 8

■♥ ❬✺❪✿ ❢r♦♠ ♣❧♦ts ✐♠♣♦rt ♣❧♦t❴✇❡✐❣❤ts❴✷❞ ♣❧♦t❴✇❡✐❣❤ts❴✷❞✭❞❞❧❛✮ 8

slide-9
SLIDE 9

❖✉t❬✺❪✿

Derivatives with regards to the inputs

9

slide-10
SLIDE 10

Masked implementations

Generate masked simulation traces

■♥ ❬✻❪✿ ❢r♦♠ ❞❡♠♦ ✐♠♣♦rt ❣❡♥❴s✐♠✉❴♠❛s❦❡❞❴❞❛t❛ ★ ●❡♥❡r❛t❡ ♠❛s❦❡❞ s✐♠✉❧❛t✐♦♥ tr❛❝❡s ❢♦r ❞❡♠♦♥str❛t✐♦♥ ❞❛t❛❴tr❛✐♥✐♥❣ ❂ ❣❡♥❴s✐♠✉❴♠❛s❦❡❞❴❞❛t❛✭✮

Demonstration: high-order DDLA

■♥ ❬✼❪✿ ❢r♦♠ ❞❡♠♦ ✐♠♣♦rt ❞❡♠♦❴❞❞❧❛❴❤✐❣❤❴♦r❞❡r ❞❞❧❛ ❂ ❞❡♠♦❴❞❞❧❛❴❤✐❣❤❴♦r❞❡r✭❞❛t❛❴tr❛✐♥✐♥❣✮ ❞❞❧❛✳r✉♥✭✷✵✮ ❞❞❧❛✳❢✐❣ ❖✉t❬✼❪✿ 10

slide-11
SLIDE 11

■♥ ❬✽❪✿ ❢r♦♠ ♣❧♦ts ✐♠♣♦rt ♣❧♦t❴✇❡✐❣❤ts❴✷❞ ♣❧♦t❴✇❡✐❣❤ts❴✷❞✭❞❞❧❛✮ ❖✉t❬✽❪✿ 11

slide-12
SLIDE 12

12

slide-13
SLIDE 13

13

slide-14
SLIDE 14

14