SlideShare a Scribd company logo
London,	August	17th	2017
Deep	Learning	vs	Learning	to	Learn	
Deep	Neural	Architecture	
Graph	Neural	Architecture	
Conclusion	
Contents
Problems	of	Deep	Learning	
Deep	Learning	is	quite	manual	for	an	AI	technology:	
▪  Designing	and	tuning	models	is	5me-consuming	and	requires	experience.	
▪  Which	hyper-parameters?	
▪  OpHmizaHon	Algorithm:	SGD?	Momentum?	RMSProp?	Adam?	AdaGrad?	
▪  Network	architecture:	so	many	choices	you	can’t	even	list	them	
▪  Classic	example:	IncepHon	architecture	(state	of	the	art	2014)	
▪  Could	you	have	guessed?	What	if	other	designs	were	beVer?	
▪  In	prac5ce:	more	an	art	than	a	science,	herding	and	heurisHcs.
Learning	to	Learn:	
Reinforcement	Learning	just	got	nominated	by	MIT	TR	in	the	10	Breakthrough	
Technologies	of	2017,	like	Deep	Learning	who	was	nominated	in	2013.	
“Ar$ficial	Intelligence	=	Reinforcement	Learning	+	Deep	Learning”	
	 	David	Silver	(AlphaGo’s	inventor)	Google	DeepMind
MIT	35	under	35	2017	list		
▪  Just	published	(Aug	2017).	Previous	winners	include	Larry	Page	&	Sergey	
Brin,	Mark	Zuckerberg,	Max	Levchin	(Paypal).	
▪  Nominees	rewarded	for	having	contributed	new	ways	for	algorithms	to	do	
things	“on	their	own”,	or	“learn	to	learn”.
Deep	Learning	vs	Learning	to	Learn	
Deep	Neural	Architecture	
Graph	Neural	Architecture	
Conclusion	
Contents
Deep	RL	for	Pong:	
▪  Three	possible	acHons:	{UP,STILL,	DOWN}.		
▪  Policy	Gradient	algorithm:	has	probabiliHes	for	each	acHon.		
▪  ProbabiliHes:	sobmax	from	raw	pixels	with	a	NN.	Weights:	Random	init.	
▪  At	each	step,	decide	acHon	by	sampling	the	sobmax	probabiliHes:		
▪  Good	final	outcome	(win)	increases	the	probabiliHes	of	ALL	the	acHons	
chosen.	A	loss	decreases	all	of	them.	Gradient	descent	(or	RMSProp)	
Links:	hVps://karpathy.github.io/2016/05/31/rl/	
hVps://gist.github.com/greydanus/5036f784eec2036252e1990da21eda18
Deep	RL	for	Neural	Nets:	
▪  Recent	Google	breakthrough	(Zoph	and	Le,	2017)	uses	same	approach	
▪  A	Recurrent	Neural	Network	(RNN)	sets	the	policy	(the	“controller”).		
▪  Weights/bias	of	the	RNN	are	randomly	iniHalized.	
▪  But	instead	of	moving	a	joysHck	up	and	down	and	see	what	happens,	it	
builds	a	“child”	NN	and	sees	what	happens.	
▪  Instead	of	score	in	a	video	game,	score	is	accuracy	R	of	“child”	NN	on	out-
of-sample	validaHon	set.		
▪  Good	final	outcome	(win)	increases	the	probabiliHes	of	ALL	the	acHons	
chosen.	A	loss	decreases	all	of	them.
Deep	RL	for	Neural	Nets:	
▪  Controller:	two-layer	LSTM	with	35	hidden	units	each	
▪  Child:	mulH-layer	convoluHonal	neural	network	CNN	
▪  Possible	acHons:	Filters	in	{24,	36,	48,	64},	filter	Height	in	{1,3,5,7}	etc.	
▪  Like	in	Pong	example,	the	acHons	are	decided	sequenHally	by	sampling	the	
sobmax	probabiliHes	(à	la	np.random.choice)	for	each	feature	and	moving	
to	the	next.	This	determines	the	CNN	child	architecture.		
▪  Training:	45,000	CIFAR	images.	Accuracy	R:	5,000	validaHon	images.	
▪  Policy	Gradient	algorithm:	REINFORCE.	(But	other	choices	are	possible).	
▪  Obtained	results	on	CIFAR-10	are	state	of	the	art	with	3.65%	error	rate
Deep	RL	for	Op5mizers:	
▪  The	same	idea	can	be	applied	to	boost	Deep	Learning	opHmizers!	
▪  An	RNN	again	sets	the	policy	(the	“controller”).		
▪  But	result	is	now	an	update	rule	opHmizer:	Adam,	RMSProp,	SGD	or	other.	
▪  Instead	of	moving	a	joysHck	up	and	down	like	in	Pong	and	see	what	
happens,	controller	builds	a	“sample”	opHmizer,	uses	it	to	train	a	neural	
network	and	sees	what	happens	to	obtained	out-of-sample	accuracy.		
▪  Good	final	outcome	(win)	increases	the	probabiliHes	of	ALL	the	acHons	
chosen.	A	loss	decreases	them.	Same	approach	as	before.
Deep	RL	for	Op5mizers:	
▪  Possible	acHons:	1st	operand	in	{g,g^2,g^3,	m,	sign(g),	sign(m)	etc},	Unary	
ops	in	{x,-x,	e^x,	clip(x,10^-4)	etc.	}	and	Binary	ops	in	{x+y,	x-y,	x*y,	x,	x/y}.		
▪  Such	”grammar”	makes	it	possible	to	build	all	classical	opHmizers:	
▪  Like	in	Pong,	acHons	are	decided	sequenHally	by	sampling	the	sobmax	
probabiliHes	for	each	feature.	This	builds	a	candidate	op5mizer.	
▪  Policy	Gradient	algorithm:	TRPO.	(But	other	choices	are	possible).
Deep	RL	for	Op5mizers:	
▪  Training	on	a	small	convoluHonal	neural	network,	the	controller	has	found	
out	two	new	update	rules,	one	of	them	is:		
▪  It	generalizes:	on	a	large	model	(Wide	ResNet)	on	CIFAR-10,	beats	all	usual	
opHmizers	and	improves	accuracy	by	up	to	2%.	
▪  It	also	beats	ADAM	on	large	NLP	tasks.
Deep	Learning	vs	Learning	to	Learn	
Deep	Neural	Architecture	
Graph	Neural	Architecture	
Conclusion	
Contents
Limita5ons	of	previous	examples	
▪  Requires	A	LOT	of	samples!	For	Neural	Architecture	Search,	authors	used	
800	GPUs	for	several	weeks!	
▪  Neural	OpHmizer	Search	uses	100	CPUs	over	days	to	find	good	opHmizers.	
▪  REINFORCE	Policy	Gradient	algorithm	is	quite	sample	inefficient.	
▪  This	makes	the	systemaHc	approaches	described	above	a	”heavy	
weaponry”	not	useful	for	everyday	needs.	
▪  Can	we	do	something	that	would	sHll	provide	results	on	a	smaller	Hme	
scale?
InstaDeep		
▪  InstaDeep	has	developed	a	platorm	to	provide	users	with	an	improvement	
on	their	favourite	architectures	quickly.	
▪  Suggested	architectures	are	more	complex	(in	terms	of	graph	theory)	and	
are	manufactured	using	neural	architecture	design.	
▪  Suggested	architectures	use	fewer	or	equal	weights	than	iniHal	one	(to	
preserve	speed).	
▪  Platorm	automa5cally	looks	for	an	op5mizer	too,	with	a	search	of	
opHmizer	space,	in	addiHon	to	the	search	of	the	neural	network	
architecture	itself.
InstaDeep’s	plaYorm		
▪  Sees	neural	networks	as	a	graph	
▪  OpHmizers	too	are	a	graph	
▪  For	example,	the	graph	on	the	right	
describes	a	4-layer	neural	net	of		
respecHvely	256,	256,	128	and	10		
with	a	non-linear	Relu	funcHon	on	the	
first	layer.	
▪  The	graph	on	the	right	describes	
an	Adam	opHmizer	with	parameters	
Beta1	=	0.5	and	beta2	=	0.9	respecHvely
InstaDeep’s	plaYorm		
▪  Sevng	up	the	environment	in	a	general	way	makes	
it	possible	to	build	RL	agents	to	perform	acHons.	
▪  Compared	to	the	RNN	controller	approach,	our	agents	are		
not	limited	to	a	specific	form	but	have	more	flexibility		
and	are	less	complex.	
	
RNN	Controller 	 	 	 	InstaDeep	
Neural	search 	 	 	 	Neural	design	
Complex	to	tune 	 	 	 	Simple	to	tune	
Time	Consuming 	 	 	 	Fast	Deployment	
OpHmal	Results 	 	 	 	Improved	Results
Demo	
Fashion	Mnist:	
“The	Mnist	for	Fashion”	
▪  28x28	grayscale	pictures	
▪  60000	training	pics	
▪  10000	test	pics	
▪  10	labelled	categories	
Experiment:	
▪  Start	with	a	2	layer	Neural	Net	
▪  Target:	improve	accuracy	on	test	set
Deep	Learning	vs	Learning	to	Learn	
Deep	Neural	Architecture	
Graph	Neural	Architecture	
Conclusion	
Contents
Conclusion	
	
▪  Neural	Architecture	methods	are	about	to	redefine	Deep	Learning,	making	
design	both	more	effec5ve	and	way	faster.		
	
▪  Already,	Meta	AI	algorithms	(like	the	ones	described)	select	cells,	layers,	and	
opHmizers.	Already	they	can	beat	the	best	human-made	neural	nets.	
	
▪  TensorFlow	community	well	posiHoned	to	benefit	from	these	advances.	
	
▪  InstaDeep	is	building	its	platorm	on	top	of	TensorFlow	to	provide	Deep	RL	
soluHons	and	boost	its	customers	AI	projects.	
▪  THANK	YOU	FOR	YOUR	TIME!	
▪  CONTACT:	hello@instadeep.com

More Related Content

Similar to TensorFlow London 13.09.17 Karim Beguir

Humans in the loop: AI in open source and industry
Humans in the loop: AI in open source and industryHumans in the loop: AI in open source and industry
Humans in the loop: AI in open source and industry
Paco Nathan
 
Data Science Accelerator Program
Data Science Accelerator ProgramData Science Accelerator Program
Data Science Accelerator Program
GoDataDriven
 
GDSC_INFO_SESSION 1.pptx
GDSC_INFO_SESSION 1.pptxGDSC_INFO_SESSION 1.pptx
GDSC_INFO_SESSION 1.pptx
VASALADEVIPRADEEP
 

Similar to TensorFlow London 13.09.17 Karim Beguir (20)

Augmenting and Automating DevOps with Artificial Intelligence
Augmenting and Automating DevOps with Artificial IntelligenceAugmenting and Automating DevOps with Artificial Intelligence
Augmenting and Automating DevOps with Artificial Intelligence
 
IIPGH Webinar 1: Getting Started With Data Science
IIPGH Webinar 1: Getting Started With Data ScienceIIPGH Webinar 1: Getting Started With Data Science
IIPGH Webinar 1: Getting Started With Data Science
 
#OSSPARIS19 - Overcoming open source challenges in reinforcement learning - W...
#OSSPARIS19 - Overcoming open source challenges in reinforcement learning - W...#OSSPARIS19 - Overcoming open source challenges in reinforcement learning - W...
#OSSPARIS19 - Overcoming open source challenges in reinforcement learning - W...
 
Future 20171110 v14
Future 20171110 v14Future 20171110 v14
Future 20171110 v14
 
How to crack Big Data and Data Science roles
How to crack Big Data and Data Science rolesHow to crack Big Data and Data Science roles
How to crack Big Data and Data Science roles
 
Humans in the loop: AI in open source and industry
Humans in the loop: AI in open source and industryHumans in the loop: AI in open source and industry
Humans in the loop: AI in open source and industry
 
Data Science-Why?What?How? By Hari Prasad
Data Science-Why?What?How? By Hari PrasadData Science-Why?What?How? By Hari Prasad
Data Science-Why?What?How? By Hari Prasad
 
A WRITER IS NOT AN EXPERT, AND AN EXPERT IS NOT A WRITER
A WRITER IS NOT AN EXPERT, AND AN EXPERT IS NOT A WRITERA WRITER IS NOT AN EXPERT, AND AN EXPERT IS NOT A WRITER
A WRITER IS NOT AN EXPERT, AND AN EXPERT IS NOT A WRITER
 
Deep learning with tensorflow
Deep learning with tensorflowDeep learning with tensorflow
Deep learning with tensorflow
 
MIT Sloan: Intro to Machine Learning
MIT Sloan: Intro to Machine LearningMIT Sloan: Intro to Machine Learning
MIT Sloan: Intro to Machine Learning
 
2023-My AI Experience - Colm Dunphy.pdf
2023-My AI Experience - Colm Dunphy.pdf2023-My AI Experience - Colm Dunphy.pdf
2023-My AI Experience - Colm Dunphy.pdf
 
(Big) Data (Science) Skills
(Big) Data (Science) Skills(Big) Data (Science) Skills
(Big) Data (Science) Skills
 
Data Science Accelerator Program
Data Science Accelerator ProgramData Science Accelerator Program
Data Science Accelerator Program
 
Deep learning
Deep learningDeep learning
Deep learning
 
BangML Meetup 1: The Path to Becoming an Machine Learning Expert
BangML Meetup 1: The Path to Becoming an Machine Learning ExpertBangML Meetup 1: The Path to Becoming an Machine Learning Expert
BangML Meetup 1: The Path to Becoming an Machine Learning Expert
 
Xiangen Hu - WESST - AutoTutor, an implementation of Conversation-Based Intel...
Xiangen Hu - WESST - AutoTutor, an implementation of Conversation-Based Intel...Xiangen Hu - WESST - AutoTutor, an implementation of Conversation-Based Intel...
Xiangen Hu - WESST - AutoTutor, an implementation of Conversation-Based Intel...
 
Data-X-v3.1
Data-X-v3.1Data-X-v3.1
Data-X-v3.1
 
R vs Python vs SAS
R vs Python vs SASR vs Python vs SAS
R vs Python vs SAS
 
Data science meetup - Spiros Antonatos
Data science meetup - Spiros AntonatosData science meetup - Spiros Antonatos
Data science meetup - Spiros Antonatos
 
GDSC_INFO_SESSION 1.pptx
GDSC_INFO_SESSION 1.pptxGDSC_INFO_SESSION 1.pptx
GDSC_INFO_SESSION 1.pptx
 

More from Seldon

TensorFlow London 18: Dr Alastair Moore, Towards the use of Graphical Models ...
TensorFlow London 18: Dr Alastair Moore, Towards the use of Graphical Models ...TensorFlow London 18: Dr Alastair Moore, Towards the use of Graphical Models ...
TensorFlow London 18: Dr Alastair Moore, Towards the use of Graphical Models ...
Seldon
 
TensorFlow London 15: Find bugs in the herd with debuggable TensorFlow code
TensorFlow London 15: Find bugs in the herd with debuggable TensorFlow code TensorFlow London 15: Find bugs in the herd with debuggable TensorFlow code
TensorFlow London 15: Find bugs in the herd with debuggable TensorFlow code
Seldon
 
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
Seldon
 
Tensorflow London 13: Zbigniew Wojna 'Deep Learning for Big Scale 2D Imagery'
Tensorflow London 13: Zbigniew Wojna 'Deep Learning for Big Scale 2D Imagery'Tensorflow London 13: Zbigniew Wojna 'Deep Learning for Big Scale 2D Imagery'
Tensorflow London 13: Zbigniew Wojna 'Deep Learning for Big Scale 2D Imagery'
Seldon
 

More from Seldon (20)

CD4ML and the challenges of testing and quality in ML systems
CD4ML and the challenges of testing and quality in ML systemsCD4ML and the challenges of testing and quality in ML systems
CD4ML and the challenges of testing and quality in ML systems
 
TensorFlow London: Cutting edge generative models
TensorFlow London: Cutting edge generative modelsTensorFlow London: Cutting edge generative models
TensorFlow London: Cutting edge generative models
 
Tensorflow London: Tensorflow and Graph Recommender Networks by Yaz Santissi
Tensorflow London: Tensorflow and Graph Recommender Networks by Yaz SantissiTensorflow London: Tensorflow and Graph Recommender Networks by Yaz Santissi
Tensorflow London: Tensorflow and Graph Recommender Networks by Yaz Santissi
 
TensorFlow London: Progressive Growing of GANs for increased stability, quali...
TensorFlow London: Progressive Growing of GANs for increased stability, quali...TensorFlow London: Progressive Growing of GANs for increased stability, quali...
TensorFlow London: Progressive Growing of GANs for increased stability, quali...
 
TensorFlow London 18: Dr Daniel Martinho-Corbishley, From science to startups...
TensorFlow London 18: Dr Daniel Martinho-Corbishley, From science to startups...TensorFlow London 18: Dr Daniel Martinho-Corbishley, From science to startups...
TensorFlow London 18: Dr Daniel Martinho-Corbishley, From science to startups...
 
TensorFlow London 18: Dr Alastair Moore, Towards the use of Graphical Models ...
TensorFlow London 18: Dr Alastair Moore, Towards the use of Graphical Models ...TensorFlow London 18: Dr Alastair Moore, Towards the use of Graphical Models ...
TensorFlow London 18: Dr Alastair Moore, Towards the use of Graphical Models ...
 
Seldon: Deploying Models at Scale
Seldon: Deploying Models at ScaleSeldon: Deploying Models at Scale
Seldon: Deploying Models at Scale
 
TensorFlow London 17: How NASA Frontier Development Lab scientists use AI to ...
TensorFlow London 17: How NASA Frontier Development Lab scientists use AI to ...TensorFlow London 17: How NASA Frontier Development Lab scientists use AI to ...
TensorFlow London 17: How NASA Frontier Development Lab scientists use AI to ...
 
TensorFlow London 17: Practical Reinforcement Learning with OpenAI
TensorFlow London 17: Practical Reinforcement Learning with OpenAITensorFlow London 17: Practical Reinforcement Learning with OpenAI
TensorFlow London 17: Practical Reinforcement Learning with OpenAI
 
TensorFlow 16: Multimodal Sentiment Analysis with TensorFlow
TensorFlow 16: Multimodal Sentiment Analysis with TensorFlow TensorFlow 16: Multimodal Sentiment Analysis with TensorFlow
TensorFlow 16: Multimodal Sentiment Analysis with TensorFlow
 
TensorFlow 16: Building a Data Science Platform
TensorFlow 16: Building a Data Science Platform TensorFlow 16: Building a Data Science Platform
TensorFlow 16: Building a Data Science Platform
 
Ai in financial services
Ai in financial servicesAi in financial services
Ai in financial services
 
TensorFlow London 15: Find bugs in the herd with debuggable TensorFlow code
TensorFlow London 15: Find bugs in the herd with debuggable TensorFlow code TensorFlow London 15: Find bugs in the herd with debuggable TensorFlow code
TensorFlow London 15: Find bugs in the herd with debuggable TensorFlow code
 
TensorFlow London 14: Ben Hall 'Machine Learning Workloads with Kubernetes an...
TensorFlow London 14: Ben Hall 'Machine Learning Workloads with Kubernetes an...TensorFlow London 14: Ben Hall 'Machine Learning Workloads with Kubernetes an...
TensorFlow London 14: Ben Hall 'Machine Learning Workloads with Kubernetes an...
 
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
 
Tensorflow London 13: Zbigniew Wojna 'Deep Learning for Big Scale 2D Imagery'
Tensorflow London 13: Zbigniew Wojna 'Deep Learning for Big Scale 2D Imagery'Tensorflow London 13: Zbigniew Wojna 'Deep Learning for Big Scale 2D Imagery'
Tensorflow London 13: Zbigniew Wojna 'Deep Learning for Big Scale 2D Imagery'
 
TensorFlow London 11: Pierre Harvey Richemond 'Trends and Developments in Rei...
TensorFlow London 11: Pierre Harvey Richemond 'Trends and Developments in Rei...TensorFlow London 11: Pierre Harvey Richemond 'Trends and Developments in Rei...
TensorFlow London 11: Pierre Harvey Richemond 'Trends and Developments in Rei...
 
TensorFlow London 11: Gema Parreno 'Use Cases of TensorFlow'
TensorFlow London 11: Gema Parreno 'Use Cases of TensorFlow'TensorFlow London 11: Gema Parreno 'Use Cases of TensorFlow'
TensorFlow London 11: Gema Parreno 'Use Cases of TensorFlow'
 
Tensorflow London 12: Marcel Horstmann and Laurent Decamp 'Using TensorFlow t...
Tensorflow London 12: Marcel Horstmann and Laurent Decamp 'Using TensorFlow t...Tensorflow London 12: Marcel Horstmann and Laurent Decamp 'Using TensorFlow t...
Tensorflow London 12: Marcel Horstmann and Laurent Decamp 'Using TensorFlow t...
 
TensorFlow London 12: Oliver Gindele 'Recommender systems in Tensorflow'
TensorFlow London 12: Oliver Gindele 'Recommender systems in Tensorflow'TensorFlow London 12: Oliver Gindele 'Recommender systems in Tensorflow'
TensorFlow London 12: Oliver Gindele 'Recommender systems in Tensorflow'
 

Recently uploaded

一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单
ewymefz
 
一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单
enxupq
 
Opendatabay - Open Data Marketplace.pptx
Opendatabay - Open Data Marketplace.pptxOpendatabay - Open Data Marketplace.pptx
Opendatabay - Open Data Marketplace.pptx
Opendatabay
 
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
vcaxypu
 
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
nscud
 
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
ukgaet
 
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
vcaxypu
 
一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单
ocavb
 
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
nscud
 
Professional Data Engineer Certification Exam Guide  _  Learn  _  Google Clou...
Professional Data Engineer Certification Exam Guide  _  Learn  _  Google Clou...Professional Data Engineer Certification Exam Guide  _  Learn  _  Google Clou...
Professional Data Engineer Certification Exam Guide  _  Learn  _  Google Clou...
Domenico Conte
 
standardisation of garbhpala offhgfffghh
standardisation of garbhpala offhgfffghhstandardisation of garbhpala offhgfffghh
standardisation of garbhpala offhgfffghh
ArpitMalhotra16
 
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
ewymefz
 
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
ewymefz
 
Empowering Data Analytics Ecosystem.pptx
Empowering Data Analytics Ecosystem.pptxEmpowering Data Analytics Ecosystem.pptx
Empowering Data Analytics Ecosystem.pptx
benishzehra469
 

Recently uploaded (20)

一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单一比一原版(NYU毕业证)纽约大学毕业证成绩单
一比一原版(NYU毕业证)纽约大学毕业证成绩单
 
一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单一比一原版(YU毕业证)约克大学毕业证成绩单
一比一原版(YU毕业证)约克大学毕业证成绩单
 
Opendatabay - Open Data Marketplace.pptx
Opendatabay - Open Data Marketplace.pptxOpendatabay - Open Data Marketplace.pptx
Opendatabay - Open Data Marketplace.pptx
 
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
一比一原版(RUG毕业证)格罗宁根大学毕业证成绩单
 
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
一比一原版(CBU毕业证)不列颠海角大学毕业证成绩单
 
Jpolillo Amazon PPC - Bid Optimization Sample
Jpolillo Amazon PPC - Bid Optimization SampleJpolillo Amazon PPC - Bid Optimization Sample
Jpolillo Amazon PPC - Bid Optimization Sample
 
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
一比一原版(UVic毕业证)维多利亚大学毕业证成绩单
 
How can I successfully sell my pi coins in Philippines?
How can I successfully sell my pi coins in Philippines?How can I successfully sell my pi coins in Philippines?
How can I successfully sell my pi coins in Philippines?
 
Innovative Methods in Media and Communication Research by Sebastian Kubitschk...
Innovative Methods in Media and Communication Research by Sebastian Kubitschk...Innovative Methods in Media and Communication Research by Sebastian Kubitschk...
Innovative Methods in Media and Communication Research by Sebastian Kubitschk...
 
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
一比一原版(ArtEZ毕业证)ArtEZ艺术学院毕业证成绩单
 
tapal brand analysis PPT slide for comptetive data
tapal brand analysis PPT slide for comptetive datatapal brand analysis PPT slide for comptetive data
tapal brand analysis PPT slide for comptetive data
 
一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单一比一原版(TWU毕业证)西三一大学毕业证成绩单
一比一原版(TWU毕业证)西三一大学毕业证成绩单
 
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
一比一原版(CBU毕业证)卡普顿大学毕业证成绩单
 
Professional Data Engineer Certification Exam Guide  _  Learn  _  Google Clou...
Professional Data Engineer Certification Exam Guide  _  Learn  _  Google Clou...Professional Data Engineer Certification Exam Guide  _  Learn  _  Google Clou...
Professional Data Engineer Certification Exam Guide  _  Learn  _  Google Clou...
 
Q1’2024 Update: MYCI’s Leap Year Rebound
Q1’2024 Update: MYCI’s Leap Year ReboundQ1’2024 Update: MYCI’s Leap Year Rebound
Q1’2024 Update: MYCI’s Leap Year Rebound
 
社内勉強会資料_LLM Agents                              .
社内勉強会資料_LLM Agents                              .社内勉強会資料_LLM Agents                              .
社内勉強会資料_LLM Agents                              .
 
standardisation of garbhpala offhgfffghh
standardisation of garbhpala offhgfffghhstandardisation of garbhpala offhgfffghh
standardisation of garbhpala offhgfffghh
 
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
一比一原版(UMich毕业证)密歇根大学|安娜堡分校毕业证成绩单
 
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单
 
Empowering Data Analytics Ecosystem.pptx
Empowering Data Analytics Ecosystem.pptxEmpowering Data Analytics Ecosystem.pptx
Empowering Data Analytics Ecosystem.pptx
 

TensorFlow London 13.09.17 Karim Beguir