SlideShare ist ein Scribd-Unternehmen logo
1 von 66
sources of
bias and explanation
Alan Dix
Computational Foundry
Swansea
http://alandix.com/academic/talks/PIT-2019-bias-and-explanation/
Tiree
Tiree Tech Wave
3-7 October
Computational Foundry
Swansea University
the foundry
building
mission
community
types of algorithms …
rules and regulations
ordinary code
classic AI
machine learning and neural nets
increasing
opacity
when things go wrong – deliberate
misuse
hacking
bad use
cyberwarfare – Stuxnet, etc.
autonomous weapons
when things go wrong – well meaning
accidents
autonomous car crashes
unintended consequences
bias (gender, ethnicity)
disproportionate social effects
https://www.bbc.co.uk/sounds/play/m00017s4 (report @ 1:41:00 in)
25 years back …
warns of the danger of gender and ethnic bias in
black-box machine learning systems
gives example: database queries using ID3
offers (partial) solution: Query-by-Browsing
and even some broader heuristics
inter alia …
yes, 25 years ago!
Query-by-Browsing
creating scructable
internal representations
Query by Browsing
user chooses records of interest
 tick for those wanted
 cross for those not wanted
system infers query
web version uses rule induction
variant of Quinlan’s ID3
www.meandeviation.com/qbb
Query by Browsing
what it looks like
user asks
system to
make a query
system infers
SQL query
query results
highlighted
Query by Browsing
dual representation
query (intensional)
for precision
listing (extensional)
for understanding
Query by Browsing – how it works
examples
machine
learning
SQL query
cond
cond
decision
tree
it is not just about
being accurate
not just right
but also upright
learning
past bias in
training data
training
data
learnt
rules
objective
function
societal
bias in goals
‘best’ may
be biased
learning
past bias in
training data
training
data
learnt
rules
objective
function
societal
bias in goals
‘best’ may
be biased
algorithms reflect society
mimicking human behaviour and choices
learning
past bias in
training data
training
data
learnt
rules
objective
function
societal
bias in goals
‘best’ may
be biased
pandering to human bias
(effective outcomes?)
• dating sites using ethnicity (CHI 2018!)
• young pretty waitresses sell more drinks
• Trump (reportedly) hiding black employees at
casino when certain rich customers arrived
• BBC (& others) paying male presenters more
because they are more popular
‘good’ business
but is it good?
learning
past bias in
training data
training
data
learnt
rules
objective
function
societal
bias in goals
‘best’ may
be biased
reinforcing societal/cultural norms
at school
boys more likely to study STEM subjects
girls more likely to study humanities
so, on average, with no other information
gender is an (albeit poor) predictor
of communication skills
and engineering knowledge
as a society we choose
to use other (and better)
predictors
innate (but largely irrelevant) differences
men are (on average) larger and stronger
so gender is a Bayesian predictor of strength
this may explain gender differences in some jobs
but …
it does NOT justify employment discrimination
bias is not about
algorithmic correctness
it is about social choice
the choice of input features
often critical in
creating or controlling bias
more data not always better!
Note:
human reasoning is
poor at ignoring low quality cues
even when we have better ones
algorithms may be better?
however …
not sufficient to remove explicit indicators:
gender/ethnicity/disability/religion
potential correlating factors e.g. clothing
algorithms need to actively avoid discrimination
and how do we know our
algorithms are OK?
Not just bias
safety – e.g. autonomous cars
democracy – e.g. social media, fake news
health and well being – e.g. soft-drink adverts
social issues – e.g. credit ratings
we need to ask
Why?
algorithmic transparency
c.f. court judgment
an AIX Kitbag
AI explainability
how to make sense of
black-box machine-learning algorithms
crucial insight …
human–human explanations
rarely utterly precise or reproducible
but are
sufficient to inspire confidence and trust
white-box black-box
grey-box
creating scructable
internal representations
analysing and
understanding
from the outside
peeking within
understanding
internal representations
but … this was all evident
25 years ago
why didn’t I do more?
if it is important
not sufficient to publish
you need to transform into
publicity and policy
white-box methods
creating scructable
internal representations
WB0. choose a white box classifier!
training set
scrutable
rules
white-box
algorithm
unseen data white-box classifier outputs
WB1. black-box generation of white box
classifier
training set
scrutable
rules
black-box
algorithm
unseen data white-box classifier outputs
WB2. Adversarial examples for white-box
learning
case-base of
behaviour scrutable
rules
black-box
adversarial learning
white-box
learning
WB3. Simplification of rule set
scrutable
rules
black-box
learning
training
set
inscrutable
rules tweak
black-box methods
analysing and understanding
from the outside
BB1. exploration analysis for human
visualisation
black-box
learning
training
set
inscrutable
rules
lots of
examples
black-box
classifier
visualise
input-output
BB2. perturbation/exploration analysis for
key feature detection
black-box
learning
inscrutable
rules
randomly vary
feature values
black-box
classifier
hotspot
visualisation
BB3. perturbation analysis for central and
boundary cases
lots of
examples
black-box
classifier
central and
boundary
cases
user
visualisation
white-box
learning
BB3. close up
central cases
perturbations
do not change class
boundary cases
small perturbations
change class
penumbra
larger perturbations
change class
BB4. black-box oracle – white-box learning
input
examples
black-box
classifier
scrutable
rules
white-box
learning
input–output
pairs as
training set
output
classes
grey-box methods
peeking within
GB0a. sensitivity analysis – weights
perturb parameters in
the inscrutable rules
lots of
examples
black-box
classifier
hotspot analysis
on parameters
GB0b. sensitivity analysis – activation
input
example
black-box classifier
(low level)
extract
intermediate
activation
black-box classifier
(high level)
perturb
activations
hotspot analysis
of nodes
GB0c. sensitivity analysis – algorithmic
apply black-box
algorithm
inverse
algorithm
GB1. high level model generation
input
examples
black-box
classifier
extract
intermediate
activation
scrutable
rules
white-box
learning
activations with
output class
as training set
output
classes
GB2. Clustering and comprehension of
low level
input
examples
black-box
classifier
extract
intermediate
activation
clusters
various
algorithms
activations
as input
MDS
SOM
GB3. triad distinctions
input
examples
black-box classifier
(low level)
A
B
C
hotspot analysis
of nodes
compare
GB4. apply generatively
output to input
activation to input
output to activation
between layers

Weitere ähnliche Inhalte

Was ist angesagt?

Formal 5 – Dialogue models – what to do when
Formal 5 – Dialogue models – what to do whenFormal 5 – Dialogue models – what to do when
Formal 5 – Dialogue models – what to do whenAlan Dix
 
What Is Interaction Design
What Is Interaction DesignWhat Is Interaction Design
What Is Interaction DesignGraeme Smith
 
HCI 3e - Ch 4 (extra):
HCI 3e - Ch 4 (extra):HCI 3e - Ch 4 (extra):
HCI 3e - Ch 4 (extra):Alan Dix
 
Human Computer Interaction (HCI)
Human Computer Interaction (HCI)Human Computer Interaction (HCI)
Human Computer Interaction (HCI)Lahiru Danushka
 
HCI 3e - Ch 19: Groupware
HCI 3e - Ch 19:  GroupwareHCI 3e - Ch 19:  Groupware
HCI 3e - Ch 19: GroupwareAlan Dix
 
Formal 6 – A success story!
Formal 6 – A success story!Formal 6 – A success story!
Formal 6 – A success story!Alan Dix
 
HCI - Chapter 4
HCI - Chapter 4HCI - Chapter 4
HCI - Chapter 4Alan Dix
 
HCI 3e - Ch 3 (extra):
HCI 3e - Ch 3 (extra):HCI 3e - Ch 3 (extra):
HCI 3e - Ch 3 (extra):Alan Dix
 
Modelling interactions: digital and physical – Part 1 – lightning tour
Modelling interactions: digital and physical – Part 1 – lightning tourModelling interactions: digital and physical – Part 1 – lightning tour
Modelling interactions: digital and physical – Part 1 – lightning tourAlan Dix
 
Designing to be used adoption appropriation
Designing to be used adoption appropriationDesigning to be used adoption appropriation
Designing to be used adoption appropriationAlan Dix
 
Designing User Interactions with AI: Servant, Master or Symbiosis.
Designing User Interactions with AI: Servant, Master or Symbiosis. Designing User Interactions with AI: Servant, Master or Symbiosis.
Designing User Interactions with AI: Servant, Master or Symbiosis. Alan Dix
 
Cognitive Engineering and User Centered Design
Cognitive Engineering and User Centered DesignCognitive Engineering and User Centered Design
Cognitive Engineering and User Centered DesignUTFPR
 
HCI 3e - Ch 13: Socio-organizational issues and stakeholder requirements
HCI 3e - Ch 13:  Socio-organizational issues and stakeholder requirementsHCI 3e - Ch 13:  Socio-organizational issues and stakeholder requirements
HCI 3e - Ch 13: Socio-organizational issues and stakeholder requirementsAlan Dix
 
HCI - Chapter 2
HCI - Chapter 2HCI - Chapter 2
HCI - Chapter 2Alan Dix
 

Was ist angesagt? (20)

Formal 5 – Dialogue models – what to do when
Formal 5 – Dialogue models – what to do whenFormal 5 – Dialogue models – what to do when
Formal 5 – Dialogue models – what to do when
 
What Is Interaction Design
What Is Interaction DesignWhat Is Interaction Design
What Is Interaction Design
 
HCI 3e - Ch 4 (extra):
HCI 3e - Ch 4 (extra):HCI 3e - Ch 4 (extra):
HCI 3e - Ch 4 (extra):
 
Human Computer Interaction (HCI)
Human Computer Interaction (HCI)Human Computer Interaction (HCI)
Human Computer Interaction (HCI)
 
Hci md exam
Hci md examHci md exam
Hci md exam
 
HCI 3e - Ch 19: Groupware
HCI 3e - Ch 19:  GroupwareHCI 3e - Ch 19:  Groupware
HCI 3e - Ch 19: Groupware
 
Formal 6 – A success story!
Formal 6 – A success story!Formal 6 – A success story!
Formal 6 – A success story!
 
HCI - Chapter 4
HCI - Chapter 4HCI - Chapter 4
HCI - Chapter 4
 
HCI 3e - Ch 3 (extra):
HCI 3e - Ch 3 (extra):HCI 3e - Ch 3 (extra):
HCI 3e - Ch 3 (extra):
 
Modelling interactions: digital and physical – Part 1 – lightning tour
Modelling interactions: digital and physical – Part 1 – lightning tourModelling interactions: digital and physical – Part 1 – lightning tour
Modelling interactions: digital and physical – Part 1 – lightning tour
 
Chapter 2
Chapter 2Chapter 2
Chapter 2
 
Hci activity#1
Hci activity#1Hci activity#1
Hci activity#1
 
Hci activity#2
Hci activity#2Hci activity#2
Hci activity#2
 
Designing to be used adoption appropriation
Designing to be used adoption appropriationDesigning to be used adoption appropriation
Designing to be used adoption appropriation
 
HCI
HCI HCI
HCI
 
Designing User Interactions with AI: Servant, Master or Symbiosis.
Designing User Interactions with AI: Servant, Master or Symbiosis. Designing User Interactions with AI: Servant, Master or Symbiosis.
Designing User Interactions with AI: Servant, Master or Symbiosis.
 
Cognitive Engineering and User Centered Design
Cognitive Engineering and User Centered DesignCognitive Engineering and User Centered Design
Cognitive Engineering and User Centered Design
 
HCI 3e - Ch 13: Socio-organizational issues and stakeholder requirements
HCI 3e - Ch 13:  Socio-organizational issues and stakeholder requirementsHCI 3e - Ch 13:  Socio-organizational issues and stakeholder requirements
HCI 3e - Ch 13: Socio-organizational issues and stakeholder requirements
 
HCI - Chapter 2
HCI - Chapter 2HCI - Chapter 2
HCI - Chapter 2
 
interaction norman model in Human Computer Interaction(HCI)
interaction  norman model in Human Computer Interaction(HCI)interaction  norman model in Human Computer Interaction(HCI)
interaction norman model in Human Computer Interaction(HCI)
 

Ähnlich wie Sources of Bias and Explanation

Sufficient Reason
Sufficient ReasonSufficient Reason
Sufficient ReasonAlan Dix
 
AI/Data Analytics (AIDA): Key concepts, examples & risks
AI/Data Analytics (AIDA): Key concepts, examples & risksAI/Data Analytics (AIDA): Key concepts, examples & risks
AI/Data Analytics (AIDA): Key concepts, examples & risksSimon Buckingham Shum
 
Applied AI Workshop - Presentation - Connect Day GDL
Applied AI Workshop - Presentation - Connect Day GDLApplied AI Workshop - Presentation - Connect Day GDL
Applied AI Workshop - Presentation - Connect Day GDLMarc Teunis
 
Interacting with an Inferred World: the Challenge of Machine Learning for Hum...
Interacting with an Inferred World: the Challenge of Machine Learning for Hum...Interacting with an Inferred World: the Challenge of Machine Learning for Hum...
Interacting with an Inferred World: the Challenge of Machine Learning for Hum...Minjoon Kim
 
Ethical Issues in Machine Learning Algorithms (Part 2)
Ethical Issues in Machine Learning Algorithms (Part 2)Ethical Issues in Machine Learning Algorithms (Part 2)
Ethical Issues in Machine Learning Algorithms (Part 2)Vladimir Kanchev
 
Responsible AI in Industry (ICML 2021 Tutorial)
Responsible AI in Industry (ICML 2021 Tutorial)Responsible AI in Industry (ICML 2021 Tutorial)
Responsible AI in Industry (ICML 2021 Tutorial)Krishnaram Kenthapadi
 
Black Box Learning Analytics? Beyond Algorithmic Transparency
Black Box Learning Analytics? Beyond Algorithmic TransparencyBlack Box Learning Analytics? Beyond Algorithmic Transparency
Black Box Learning Analytics? Beyond Algorithmic TransparencySimon Buckingham Shum
 
M2 l10 fairness, accountability, and transparency
M2 l10 fairness, accountability, and transparencyM2 l10 fairness, accountability, and transparency
M2 l10 fairness, accountability, and transparencyBoPeng76
 
Artificial Intelligence
Artificial Intelligence Artificial Intelligence
Artificial Intelligence Prasad Kulkarni
 
Real World NLP, ML, and Big Data
Real World NLP, ML, and Big DataReal World NLP, ML, and Big Data
Real World NLP, ML, and Big DataDevin Bost
 
Re-Empower the Public with Data Visualization and Game Design
Re-Empower the Public with Data Visualization and Game DesignRe-Empower the Public with Data Visualization and Game Design
Re-Empower the Public with Data Visualization and Game DesignSam Pottinger
 
Information Architecture for Retail Web Sites: Lessons from the Field
Information Architecture for Retail Web Sites: Lessons from the FieldInformation Architecture for Retail Web Sites: Lessons from the Field
Information Architecture for Retail Web Sites: Lessons from the FieldNick Berry
 
Summer@UTS: Visualisation for Wicked Problems
Summer@UTS: Visualisation for Wicked ProblemsSummer@UTS: Visualisation for Wicked Problems
Summer@UTS: Visualisation for Wicked ProblemsSimon Buckingham Shum
 
E3 Chap 05 Interaction Design Basics
E3 Chap 05 Interaction Design BasicsE3 Chap 05 Interaction Design Basics
E3 Chap 05 Interaction Design BasicsGameo
 
Writing A Thesis Statement For Resea
Writing A Thesis Statement For ReseaWriting A Thesis Statement For Resea
Writing A Thesis Statement For ReseaJennifer Strong
 
Designing A.I. - Week 1 - Intro Lecture
Designing A.I. - Week 1 - Intro LectureDesigning A.I. - Week 1 - Intro Lecture
Designing A.I. - Week 1 - Intro LectureDavid Young
 
Explainable AI in Industry (AAAI 2020 Tutorial)
Explainable AI in Industry (AAAI 2020 Tutorial)Explainable AI in Industry (AAAI 2020 Tutorial)
Explainable AI in Industry (AAAI 2020 Tutorial)Krishnaram Kenthapadi
 
Joachim Ganseman - Pitfalls in AI - Infosecurity.be 2019
Joachim Ganseman - Pitfalls in AI - Infosecurity.be 2019Joachim Ganseman - Pitfalls in AI - Infosecurity.be 2019
Joachim Ganseman - Pitfalls in AI - Infosecurity.be 2019Smals
 
Human in the loop: Bayesian Rules Enabling Explainable AI
Human in the loop: Bayesian Rules Enabling Explainable AIHuman in the loop: Bayesian Rules Enabling Explainable AI
Human in the loop: Bayesian Rules Enabling Explainable AIPramit Choudhary
 
02 Network Data Collection
02 Network Data Collection02 Network Data Collection
02 Network Data Collectiondnac
 

Ähnlich wie Sources of Bias and Explanation (20)

Sufficient Reason
Sufficient ReasonSufficient Reason
Sufficient Reason
 
AI/Data Analytics (AIDA): Key concepts, examples & risks
AI/Data Analytics (AIDA): Key concepts, examples & risksAI/Data Analytics (AIDA): Key concepts, examples & risks
AI/Data Analytics (AIDA): Key concepts, examples & risks
 
Applied AI Workshop - Presentation - Connect Day GDL
Applied AI Workshop - Presentation - Connect Day GDLApplied AI Workshop - Presentation - Connect Day GDL
Applied AI Workshop - Presentation - Connect Day GDL
 
Interacting with an Inferred World: the Challenge of Machine Learning for Hum...
Interacting with an Inferred World: the Challenge of Machine Learning for Hum...Interacting with an Inferred World: the Challenge of Machine Learning for Hum...
Interacting with an Inferred World: the Challenge of Machine Learning for Hum...
 
Ethical Issues in Machine Learning Algorithms (Part 2)
Ethical Issues in Machine Learning Algorithms (Part 2)Ethical Issues in Machine Learning Algorithms (Part 2)
Ethical Issues in Machine Learning Algorithms (Part 2)
 
Responsible AI in Industry (ICML 2021 Tutorial)
Responsible AI in Industry (ICML 2021 Tutorial)Responsible AI in Industry (ICML 2021 Tutorial)
Responsible AI in Industry (ICML 2021 Tutorial)
 
Black Box Learning Analytics? Beyond Algorithmic Transparency
Black Box Learning Analytics? Beyond Algorithmic TransparencyBlack Box Learning Analytics? Beyond Algorithmic Transparency
Black Box Learning Analytics? Beyond Algorithmic Transparency
 
M2 l10 fairness, accountability, and transparency
M2 l10 fairness, accountability, and transparencyM2 l10 fairness, accountability, and transparency
M2 l10 fairness, accountability, and transparency
 
Artificial Intelligence
Artificial Intelligence Artificial Intelligence
Artificial Intelligence
 
Real World NLP, ML, and Big Data
Real World NLP, ML, and Big DataReal World NLP, ML, and Big Data
Real World NLP, ML, and Big Data
 
Re-Empower the Public with Data Visualization and Game Design
Re-Empower the Public with Data Visualization and Game DesignRe-Empower the Public with Data Visualization and Game Design
Re-Empower the Public with Data Visualization and Game Design
 
Information Architecture for Retail Web Sites: Lessons from the Field
Information Architecture for Retail Web Sites: Lessons from the FieldInformation Architecture for Retail Web Sites: Lessons from the Field
Information Architecture for Retail Web Sites: Lessons from the Field
 
Summer@UTS: Visualisation for Wicked Problems
Summer@UTS: Visualisation for Wicked ProblemsSummer@UTS: Visualisation for Wicked Problems
Summer@UTS: Visualisation for Wicked Problems
 
E3 Chap 05 Interaction Design Basics
E3 Chap 05 Interaction Design BasicsE3 Chap 05 Interaction Design Basics
E3 Chap 05 Interaction Design Basics
 
Writing A Thesis Statement For Resea
Writing A Thesis Statement For ReseaWriting A Thesis Statement For Resea
Writing A Thesis Statement For Resea
 
Designing A.I. - Week 1 - Intro Lecture
Designing A.I. - Week 1 - Intro LectureDesigning A.I. - Week 1 - Intro Lecture
Designing A.I. - Week 1 - Intro Lecture
 
Explainable AI in Industry (AAAI 2020 Tutorial)
Explainable AI in Industry (AAAI 2020 Tutorial)Explainable AI in Industry (AAAI 2020 Tutorial)
Explainable AI in Industry (AAAI 2020 Tutorial)
 
Joachim Ganseman - Pitfalls in AI - Infosecurity.be 2019
Joachim Ganseman - Pitfalls in AI - Infosecurity.be 2019Joachim Ganseman - Pitfalls in AI - Infosecurity.be 2019
Joachim Ganseman - Pitfalls in AI - Infosecurity.be 2019
 
Human in the loop: Bayesian Rules Enabling Explainable AI
Human in the loop: Bayesian Rules Enabling Explainable AIHuman in the loop: Bayesian Rules Enabling Explainable AI
Human in the loop: Bayesian Rules Enabling Explainable AI
 
02 Network Data Collection
02 Network Data Collection02 Network Data Collection
02 Network Data Collection
 

Mehr von Alan Dix

CDT Away Day Talk: Qualitative–Quantitative reasoning and lightweight numbers
CDT Away Day Talk: Qualitative–Quantitative reasoning and lightweight numbersCDT Away Day Talk: Qualitative–Quantitative reasoning and lightweight numbers
CDT Away Day Talk: Qualitative–Quantitative reasoning and lightweight numbersAlan Dix
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Alan Dix
 
From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .Alan Dix
 
Human-Centred Artificial Intelligence – Malta 2024
Human-Centred Artificial Intelligence – Malta 2024Human-Centred Artificial Intelligence – Malta 2024
Human-Centred Artificial Intelligence – Malta 2024Alan Dix
 
The future of UX design support tools - talk Paris March 2024
The future of UX design support tools - talk Paris March 2024The future of UX design support tools - talk Paris March 2024
The future of UX design support tools - talk Paris March 2024Alan Dix
 
Qualitative–Quantitative reasoning and lightweight numbers
Qualitative–Quantitative reasoning and lightweight numbersQualitative–Quantitative reasoning and lightweight numbers
Qualitative–Quantitative reasoning and lightweight numbersAlan Dix
 
Invited talk at Diversifying Knowledge Production in HCI
Invited talk at Diversifying Knowledge Production in HCIInvited talk at Diversifying Knowledge Production in HCI
Invited talk at Diversifying Knowledge Production in HCIAlan Dix
 
Exceptional Experiences for Everyone
Exceptional Experiences for EveryoneExceptional Experiences for Everyone
Exceptional Experiences for EveryoneAlan Dix
 
Inclusivity and AI: opportunity or threat
Inclusivity and AI: opportunity or threatInclusivity and AI: opportunity or threat
Inclusivity and AI: opportunity or threatAlan Dix
 
Hidden Figures architectural challenges to expose parameters lost in code
Hidden Figures architectural challenges to expose parameters lost in codeHidden Figures architectural challenges to expose parameters lost in code
Hidden Figures architectural challenges to expose parameters lost in codeAlan Dix
 
ChatGPT, Culture and Creativity simulacrum and alterity
ChatGPT, Culture and Creativity simulacrum and alterityChatGPT, Culture and Creativity simulacrum and alterity
ChatGPT, Culture and Creativity simulacrum and alterityAlan Dix
 
Why pandemics and climate change are hard to understand and make decision mak...
Why pandemics and climate change are hard to understand and make decision mak...Why pandemics and climate change are hard to understand and make decision mak...
Why pandemics and climate change are hard to understand and make decision mak...Alan Dix
 
Beyond the Wireframe: tools to design, analyse and prototype physical devices
Beyond the Wireframe: tools to design, analyse and prototype physical devicesBeyond the Wireframe: tools to design, analyse and prototype physical devices
Beyond the Wireframe: tools to design, analyse and prototype physical devicesAlan Dix
 
Forever Cyborgs – a long view on physical-digital interaction
Forever Cyborgs – a long view on physical-digital interactionForever Cyborgs – a long view on physical-digital interaction
Forever Cyborgs – a long view on physical-digital interactionAlan Dix
 
Truth in an Age of Information
Truth in an Age of InformationTruth in an Age of Information
Truth in an Age of InformationAlan Dix
 
Rome Seminar: Designing User Interactions with AI
Rome Seminar: Designing User Interactions with AIRome Seminar: Designing User Interactions with AI
Rome Seminar: Designing User Interactions with AIAlan Dix
 
Tools and technology to support rich community heritage
Tools and technology to support rich community heritageTools and technology to support rich community heritage
Tools and technology to support rich community heritageAlan Dix
 
Maps with Meaning
Maps with MeaningMaps with Meaning
Maps with MeaningAlan Dix
 
Democratising Digitisation Tools to Support Small Community Archives
Democratising Digitisation Tools to Support Small Community ArchivesDemocratising Digitisation Tools to Support Small Community Archives
Democratising Digitisation Tools to Support Small Community ArchivesAlan Dix
 
Follow your nose: history frames the future
Follow your nose: history frames the futureFollow your nose: history frames the future
Follow your nose: history frames the futureAlan Dix
 

Mehr von Alan Dix (20)

CDT Away Day Talk: Qualitative–Quantitative reasoning and lightweight numbers
CDT Away Day Talk: Qualitative–Quantitative reasoning and lightweight numbersCDT Away Day Talk: Qualitative–Quantitative reasoning and lightweight numbers
CDT Away Day Talk: Qualitative–Quantitative reasoning and lightweight numbers
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
 
From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .
 
Human-Centred Artificial Intelligence – Malta 2024
Human-Centred Artificial Intelligence – Malta 2024Human-Centred Artificial Intelligence – Malta 2024
Human-Centred Artificial Intelligence – Malta 2024
 
The future of UX design support tools - talk Paris March 2024
The future of UX design support tools - talk Paris March 2024The future of UX design support tools - talk Paris March 2024
The future of UX design support tools - talk Paris March 2024
 
Qualitative–Quantitative reasoning and lightweight numbers
Qualitative–Quantitative reasoning and lightweight numbersQualitative–Quantitative reasoning and lightweight numbers
Qualitative–Quantitative reasoning and lightweight numbers
 
Invited talk at Diversifying Knowledge Production in HCI
Invited talk at Diversifying Knowledge Production in HCIInvited talk at Diversifying Knowledge Production in HCI
Invited talk at Diversifying Knowledge Production in HCI
 
Exceptional Experiences for Everyone
Exceptional Experiences for EveryoneExceptional Experiences for Everyone
Exceptional Experiences for Everyone
 
Inclusivity and AI: opportunity or threat
Inclusivity and AI: opportunity or threatInclusivity and AI: opportunity or threat
Inclusivity and AI: opportunity or threat
 
Hidden Figures architectural challenges to expose parameters lost in code
Hidden Figures architectural challenges to expose parameters lost in codeHidden Figures architectural challenges to expose parameters lost in code
Hidden Figures architectural challenges to expose parameters lost in code
 
ChatGPT, Culture and Creativity simulacrum and alterity
ChatGPT, Culture and Creativity simulacrum and alterityChatGPT, Culture and Creativity simulacrum and alterity
ChatGPT, Culture and Creativity simulacrum and alterity
 
Why pandemics and climate change are hard to understand and make decision mak...
Why pandemics and climate change are hard to understand and make decision mak...Why pandemics and climate change are hard to understand and make decision mak...
Why pandemics and climate change are hard to understand and make decision mak...
 
Beyond the Wireframe: tools to design, analyse and prototype physical devices
Beyond the Wireframe: tools to design, analyse and prototype physical devicesBeyond the Wireframe: tools to design, analyse and prototype physical devices
Beyond the Wireframe: tools to design, analyse and prototype physical devices
 
Forever Cyborgs – a long view on physical-digital interaction
Forever Cyborgs – a long view on physical-digital interactionForever Cyborgs – a long view on physical-digital interaction
Forever Cyborgs – a long view on physical-digital interaction
 
Truth in an Age of Information
Truth in an Age of InformationTruth in an Age of Information
Truth in an Age of Information
 
Rome Seminar: Designing User Interactions with AI
Rome Seminar: Designing User Interactions with AIRome Seminar: Designing User Interactions with AI
Rome Seminar: Designing User Interactions with AI
 
Tools and technology to support rich community heritage
Tools and technology to support rich community heritageTools and technology to support rich community heritage
Tools and technology to support rich community heritage
 
Maps with Meaning
Maps with MeaningMaps with Meaning
Maps with Meaning
 
Democratising Digitisation Tools to Support Small Community Archives
Democratising Digitisation Tools to Support Small Community ArchivesDemocratising Digitisation Tools to Support Small Community Archives
Democratising Digitisation Tools to Support Small Community Archives
 
Follow your nose: history frames the future
Follow your nose: history frames the futureFollow your nose: history frames the future
Follow your nose: history frames the future
 

Kürzlich hochgeladen

A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024The Digital Insurer
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024The Digital Insurer
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAndrey Devyatkin
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘RTylerCroy
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century educationjfdjdjcjdnsjd
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Miguel Araújo
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherRemote DBA Services
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Drew Madelung
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Scriptwesley chun
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
HTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesHTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesBoston Institute of Analytics
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 

Kürzlich hochgeladen (20)

A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
HTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesHTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation Strategies
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 

Sources of Bias and Explanation

  • 1. sources of bias and explanation Alan Dix Computational Foundry Swansea http://alandix.com/academic/talks/PIT-2019-bias-and-explanation/
  • 2. Tiree Tiree Tech Wave 3-7 October Computational Foundry Swansea University
  • 4.
  • 5.
  • 6.
  • 7. types of algorithms … rules and regulations ordinary code classic AI machine learning and neural nets increasing opacity
  • 8. when things go wrong – deliberate misuse hacking bad use cyberwarfare – Stuxnet, etc. autonomous weapons
  • 9. when things go wrong – well meaning accidents autonomous car crashes unintended consequences bias (gender, ethnicity) disproportionate social effects https://www.bbc.co.uk/sounds/play/m00017s4 (report @ 1:41:00 in)
  • 11.
  • 12. warns of the danger of gender and ethnic bias in black-box machine learning systems gives example: database queries using ID3 offers (partial) solution: Query-by-Browsing and even some broader heuristics inter alia …
  • 14.
  • 16. Query by Browsing user chooses records of interest  tick for those wanted  cross for those not wanted system infers query web version uses rule induction variant of Quinlan’s ID3 www.meandeviation.com/qbb
  • 17. Query by Browsing what it looks like user asks system to make a query system infers SQL query query results highlighted
  • 18. Query by Browsing dual representation query (intensional) for precision listing (extensional) for understanding
  • 19. Query by Browsing – how it works examples machine learning SQL query cond cond decision tree
  • 20.
  • 21. it is not just about being accurate not just right but also upright
  • 22. learning past bias in training data training data learnt rules objective function societal bias in goals ‘best’ may be biased
  • 23. learning past bias in training data training data learnt rules objective function societal bias in goals ‘best’ may be biased
  • 26. learning past bias in training data training data learnt rules objective function societal bias in goals ‘best’ may be biased
  • 27. pandering to human bias (effective outcomes?) • dating sites using ethnicity (CHI 2018!) • young pretty waitresses sell more drinks • Trump (reportedly) hiding black employees at casino when certain rich customers arrived • BBC (& others) paying male presenters more because they are more popular
  • 29. learning past bias in training data training data learnt rules objective function societal bias in goals ‘best’ may be biased
  • 30. reinforcing societal/cultural norms at school boys more likely to study STEM subjects girls more likely to study humanities so, on average, with no other information gender is an (albeit poor) predictor of communication skills and engineering knowledge
  • 31. as a society we choose to use other (and better) predictors
  • 32. innate (but largely irrelevant) differences men are (on average) larger and stronger so gender is a Bayesian predictor of strength this may explain gender differences in some jobs but … it does NOT justify employment discrimination
  • 33. bias is not about algorithmic correctness it is about social choice
  • 34. the choice of input features often critical in creating or controlling bias more data not always better!
  • 35. Note: human reasoning is poor at ignoring low quality cues even when we have better ones
  • 36. algorithms may be better?
  • 37. however … not sufficient to remove explicit indicators: gender/ethnicity/disability/religion potential correlating factors e.g. clothing algorithms need to actively avoid discrimination
  • 38. and how do we know our algorithms are OK?
  • 39. Not just bias safety – e.g. autonomous cars democracy – e.g. social media, fake news health and well being – e.g. soft-drink adverts social issues – e.g. credit ratings
  • 40. we need to ask Why? algorithmic transparency c.f. court judgment
  • 41.
  • 42. an AIX Kitbag AI explainability how to make sense of black-box machine-learning algorithms
  • 43. crucial insight … human–human explanations rarely utterly precise or reproducible but are sufficient to inspire confidence and trust
  • 44. white-box black-box grey-box creating scructable internal representations analysing and understanding from the outside peeking within understanding internal representations
  • 45.
  • 46. but … this was all evident 25 years ago why didn’t I do more? if it is important not sufficient to publish you need to transform into publicity and policy
  • 47.
  • 49. WB0. choose a white box classifier! training set scrutable rules white-box algorithm unseen data white-box classifier outputs
  • 50. WB1. black-box generation of white box classifier training set scrutable rules black-box algorithm unseen data white-box classifier outputs
  • 51. WB2. Adversarial examples for white-box learning case-base of behaviour scrutable rules black-box adversarial learning white-box learning
  • 52. WB3. Simplification of rule set scrutable rules black-box learning training set inscrutable rules tweak
  • 53. black-box methods analysing and understanding from the outside
  • 54. BB1. exploration analysis for human visualisation black-box learning training set inscrutable rules lots of examples black-box classifier visualise input-output
  • 55. BB2. perturbation/exploration analysis for key feature detection black-box learning inscrutable rules randomly vary feature values black-box classifier hotspot visualisation
  • 56. BB3. perturbation analysis for central and boundary cases lots of examples black-box classifier central and boundary cases user visualisation white-box learning
  • 57. BB3. close up central cases perturbations do not change class boundary cases small perturbations change class penumbra larger perturbations change class
  • 58. BB4. black-box oracle – white-box learning input examples black-box classifier scrutable rules white-box learning input–output pairs as training set output classes
  • 60. GB0a. sensitivity analysis – weights perturb parameters in the inscrutable rules lots of examples black-box classifier hotspot analysis on parameters
  • 61. GB0b. sensitivity analysis – activation input example black-box classifier (low level) extract intermediate activation black-box classifier (high level) perturb activations hotspot analysis of nodes
  • 62. GB0c. sensitivity analysis – algorithmic apply black-box algorithm inverse algorithm
  • 63. GB1. high level model generation input examples black-box classifier extract intermediate activation scrutable rules white-box learning activations with output class as training set output classes
  • 64. GB2. Clustering and comprehension of low level input examples black-box classifier extract intermediate activation clusters various algorithms activations as input MDS SOM
  • 65. GB3. triad distinctions input examples black-box classifier (low level) A B C hotspot analysis of nodes compare
  • 66. GB4. apply generatively output to input activation to input output to activation between layers