Machine Learning meets Blockchain

Apr 7, 20190 likes291 views

Machine learning and blockchain each have their own strengths that could be combined to create more powerful applications. Machine learning excels at making predictions from large datasets but relies on centralized systems and lacks transparency. Blockchain provides a decentralized, transparent system for recording transactions but does not have machine learning's predictive capabilities. By combining the two, it may be possible to develop models with better performance that are also securely deployed on decentralized networks using blockchain. This could result in more data leading to better models, more accurate results and improved decisions over time.

What Happens If
Machine Learning
meets Blockchain
Sefik Ilkin Serengil

sefiks.com
serengil
Sefik Ilkin Serengil
Software Developer @ ‘10
CS graduate, MSc @ ’11

Data is the new oil (The Economist, 2017)
753B 739B72B782B 138B 2.2B
Total: 2.4T $ France GDP: 2.5T $

Geoffrey Hinton
Professor at University of Toronto, Fellow at Google Brain
Inventor of Backpropagation - Learning Algorithm of Neural Networks
Deep Learning is an algorithm which has no
theoretical limitations of what it can learn; the
more data you give and the more
computational time you provide, the better it is

Andrew Ng
Prof. at Stanford University
Founder of Coursera, Chief Scientist at Baidu
I have worked all my life in Machine
Learning, and I’ve never seen one
algorithm knock over benchmarks like
Deep Learning

robertaguileradesign.com
Fool your
neural networks

Reported to have lost $136.5 billion in market cap

https://towardsdatascience.com/breaking-neural-networks-with-adversarial-attacks-f4290a9a45aa
Classified as Speed Limit: 45 mph sign!

Peter Thiel
Founder of Paypal
Angel investor of Facebook
Crypto is libertarian,
AI is communist.
https://www.youtube.com/watch?v=-iYpmOV60ks

Reid Hoffman
Cofounder of Linkedin and Paypal
Angel investor of Facebook
Crypto is anarchy,
AI is the rule of laws
https://www.youtube.com/watch?v=-iYpmOV60ks

Blockchain is deterministic whereas ML is probabilistic

Blockchain has a decentralized architecture

Deployment Time
BusinessValue
Model
Model
Remodel / retrain
ML models are changing
Model
PositiveNegative

Secure &
More Data
Better
Models
Better
Results
Better
Actions

sefiks.com serengil
Thank you for your attention!
Credit: Yasemin Dalkilic

Deep learning is the fastest growing field in artificial intelligence according to Gartner's top strategic technology trends for 2017 which ranks AI and machine learning as number one. The document then provides examples of how deep learning is accelerating innovation in various industries such as cancer research, speech recognition, and more. It promotes NVIDIA's deep learning and AI solutions to help readers learn how these technologies can impact their own business.

Challenges in Building Operational AI - Daniel Skantze at Jfokus 2019 Peltarion

��

This document discusses challenges in building operational AI systems and provides guidance on how to approach AI development. It notes that while AI will significantly impact many aspects of life, building reliable operational AI systems faces hurdles such as a lack of high-quality data, problems being underspecified, tools not being designed for real-world use, and proof-of-concept systems failing to scale. The document offers design guidelines for AI such as ensuring ecological validity, being data-driven, making all aspects immutable and version-controlled, considering scaling from the start, and avoiding strong dependencies. It also provides advice on getting started with AI through self-study, using existing solutions, implementing non-critical test cases, and growing knowledge and processes

Top Data Analysts in the world you can find at XpertAravind kumar

��

Artificial Intelligence Applications, Research, and EconomicsIkhlaq Sidhu

��

Ikhlaq Sidhu is the Founding Faculty Director of the Sutardja Center for Entrepreneurship & Technology at UC Berkeley. He discusses artificial intelligence applications and research at Berkeley, as well as perspectives on the technical evolution of AI/ML. Regarding job loss concerns from AI, Sidhu notes that past industrial revolutions ultimately created more jobs despite initial disruption, and that retraining will be essential for workers to transition. Sidhu emphasizes that a "multiplicity" scenario is more likely than a singular takeover by AI, and that humans will continue playing important roles as technologies are developed and integrated.

AI for All: Biology is eating the world & AI is eating Biology Intel® Software

��

Advances in cell biology and creation of an immense amount of data are converging with advances in Machine learning to analyze this data. Biology is experiencing its AI moment and driving the massive computation involved in understanding biological mechanisms and driving interventions. Learn about how cutting edge technologies such as Software Guard Extensions (SGX) in the latest Intel Xeon Processors and Open Federated Learning (OpenFL), an open framework for federated learning developed by Intel, are helping advance AI in gene therapy, drug design, disease identification and more.

9/16 Top 5 Deep LearningNVIDIA

��

Deep learning is the fastest growing field in artificial intelligence. Andrew Ng, Chief Scientist at Baidu Research, explains that the rise of deep learning can be attributed to two major trends: 1) increased computation power and data availability, and 2) deep learning's ability to output complex things like text, images, and audio. This transformation ability could change industries. The document then provides five examples of how AI and deep learning are accelerating innovation, including Microsoft achieving a speech recognition milestone and Houzz using deep learning for product discovery.

50th Anniversary Keynote for Korean Testing LaboratoryJerome Glenn

��

The document discusses emerging and converging future technologies like artificial intelligence, robotics, 3D printing, synthetic biology, and their synergistic effects. It argues these technologies, combined with computational science and Moore's law, will accelerate progress faster than any individual technology. The document advocates developing collective intelligence systems to help anticipate, manage and guide technological change, for example to help the Korean Testing Laboratory strategize and stay ahead of emerging opportunities. It also briefly discusses some implications of these technologies for the future of work, consciousness, and testing methods.

Hacktoberfest'24 _ GDG on Campus BU.pptxnilaygupta3003

��

Machine Learning in Cyber Security Domain BGA Cyber Security

��

The document discusses machine learning and its applications in cyber security. It provides an introduction to machine learning and how it is used to analyze large amounts of data and make decisions without being explicitly programmed. Examples of machine learning applications discussed include recommendation systems, activity recognition, weather forecasting, and image processing. The document also discusses how machine learning is being applied in cyber security to help detect sophisticated cyber attacks.

building intelligent systems with large scale deep learningmustafa sarac

��

The document discusses the work of the Google Brain team in conducting long-term research on machine learning and building systems like TensorFlow to make ML models more widely available. It outlines the team's goals of making machines intelligent to improve people's lives through research areas like computer vision, healthcare, robotics and language understanding. The team aims to build general tools for ML and collaborate within Google and with others to apply their research at large scale.

ButlerThomas Butler, CISSP, CEH, CHFI, CISA, CPA, CIA

��

This document outlines a presentation on creating a penetration testing laboratory for hands-on practice. It describes setting up three virtual machines using free software, including an "attack machine" with Backtrack5R3 and two "victim machines" with vulnerable operating systems. The project consists of seven appendices that document the lab creation, penetration testing methodology, reconnaissance, scanning, exploitation, post-exploitation, and technical terms. The presentation discusses the need for more hands-on training and the challenges of keeping penetration testing skills sharp.

DeepSeek AI Review: The AI Chatbot Shaping the Future of Conversational AIMasoomAkhtar2

��

DeepSeek: The AI Chatbot Shaping the Future of Conversational AI In today’s fast-moving world of artificial intelligence, a new contender has stepped onto the scene, challenging OpenAI’s ChatGPT's dominance. Meet DeepSeek, an advanced AI chatbot turning heads with its cutting-edge technology, efficient training process, and futuristic approach to natural language understanding. But what exactly is DeepSeek, and how does it stack up against the AI giants? Let’s break it down. For a complete post, pleas visit: https://dearteacher4u.blogspot.com/2025/02/deepseek-ai-review.html

Ethical Hacking, Its relevance and Its ProspectsRwik Kumar Dutta

��

9/30 Top 5 Deep Learning NVIDIA

��

Deep learning is the fastest growing field in artificial intelligence. Leaders in various industries believe that within five years, AI will transform many businesses in ways not seen since the computer revolution. Deep learning is accelerating innovation across industries and is changing lives through applications like image captioning and database analytics that are powered by GPUs to enable fast training and inference.

How To Become an Ethical Hacker?Srashti Jain

��

Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014Austin Ogilvie

��

The document outlines Greg Lamp's presentation at a Data Science MD Meetup in October 2014 about Applied Data Science with Yhat. The presentation covers the challenges of building analytical applications, a case study of a beer recommender system built in Python using beer review data, and a demonstration of deploying the model through Yhat's platform. It concludes with a question and answer section.

9/9/16 Top 5 Deep LearningNVIDIA

��

Deep learning is the fastest growing field in artificial intelligence. It has the potential to transform industries like electricity did 100 years ago. The document highlights five stories of how AI and deep learning are accelerating innovation: 1) Baidu open sourced its deep learning platform PaddlePaddle to attract talent, 2) Machine learning will push data science to increase relevance, 3) UC Berkeley created artificial intelligence graders to cut grading time by 75%, 4) A startup developed an algorithm to recognize objects in photos and link them to items for sale, 5) Deep learning could help football coaches with strategic insights.

Citizen AI Engineer Program 2018 CAI 500 Fast Track AI Week1 RoadmapDr. Mohan K. Bavirisetty

��

The document provides an overview and roadmap for Week 1 of the Citizen AI Engineer Program 2018. It outlines the mission to provide free and open training in artificial intelligence. Week 1 will cover an introduction to AI, recommended learning resources, social engagement activities, optional exercises and certification options. Participants can work towards Green Belt, Brown Belt or Black Belt certifications by completing modules and a final project. The roadmap recommends installing R, exploring AI libraries, joining local meetups, and analyzing breast cancer or prime number datasets.

HKUST Computer Science Festival 2013 - Seminar: Computer Science, Hacking and...Anthony Lai

��

Defining a Practical Path to Artificial Intelligence Roman Chanclor

��

The Programmable Internet of ThingsRich Miller

��

2015 Bio-IT Trends From the TrenchesChris Dagdigian

��

This document provides an overview of trends from the 2015 Bio IT World Expo presented by Chris Dag from the BioTeam. It discusses trends around DevOps, automation, converged infrastructure, compute, storage, cloud computing, and specific projects. Key points include the need for sysadmins to learn scripting and automation, the growing role of APIs, and how object storage is the future for managing scientific data and metadata. Specific examples highlight high performance computing configurations, small file ingest solutions, low-cost storage approaches, and a petascale storage system built using Intel Lustre, Linux, and ZFS on commodity hardware.

10/13 Top 5 Deep Learning StoriesNVIDIA

��

NYU's Center for Data Science recently acquired an NVIDIA DGX-1 AI supercomputer to accelerate research in fields such as machine learning, computer vision, and robotics. The director of AI Research at Facebook says the DGX-1 will be crucial for success in nearly every research project at NYU. OpenAI is using discussions on Reddit to increase the amount of natural language data available to train their AI systems to communicate like humans.

CompetitiveAdvantageThroughDeepLearning (white paper)Tarun Mehra

��

This white paper discusses deep learning and its value for businesses. It outlines four value levers for businesses to consider: math, technology, applications, and data. While deep learning is becoming mainstream, many executives do not fully understand its value levers and how to apply them strategically. The paper provides background on the history and recent developments in deep learning to help executives better understand and leverage its opportunities.

What does "monitoring" mean? (FOSDEM 2017)Brian Brazil

��

Monitoring can mean very different things to different people, and this often leads to confusion and misunderstandings. There are many offerings both free software and commercials, and it's not always clear where each fits in the bigger picture. This talk will look a bit at the history of monitoring, and then into the general categories of Metrics, Logs, Profiling and Distributed tracing and how each of these is important in Cloud-based environment. Video: https://www.youtube.com/watch?v=hCBGyLRJ1qo

Notes on Simulation and GHDLDIlawar Singh

��

This document provides an overview of simulation using GHDL for verifying digital designs. It discusses that simulation is important for detecting errors in a design early to reduce costs. Verification through simulation involves creating test benches to test a design against different input combinations. For large designs, decomposition is used to break them into smaller components that can be simulated more easily. The document advocates for using the open-source GHDL simulator, noting the benefits of understanding and improving open-source software. It provides instructions for installing GHDL on Linux and Windows and discusses using text editors like Vim or Gedit when working with GHDL. Coding best practices like commenting and readability are also mentioned.

"Large-Scale Deep Learning for Building Intelligent Computer Systems," a Keyn...Edge AI and Vision Alliance

��

For the full video of this presentation, please visit: http://www.embedded-vision.com/platinum-members/embedded-vision-alliance/embedded-vision-training/videos/pages/may-2016-embedded-vision-summit-google-keynote For more information about embedded vision, please visit: http://www.embedded-vision.com Jeff Dean, Senior Fellow at Google, presents the "Large-Scale Deep Learning for Building Intelligent Computer Systems" keynote at the May 2016 Embedded Vision Summit. Over the past few years, Google has built two generations of large-scale computer systems for training neural networks, and then applied these systems to a wide variety of research problems that have traditionally been very difficult for computers. Google has released its second generation system, TensorFlow, as an open source project, and is now collaborating with a growing community on improving and extending its functionality. Using TensorFlow, Google's research group has made significant improvements in the state-of-the-art in many areas, and dozens of different groups at Google use it to train state-of-the-art models for speech recognition, image recognition, various visual detection tasks, language modeling, language translation, and many other tasks. In this talk, Jeff highlights some of ways that Google trains large models quickly on large datasets, and discusses different approaches for deploying machine learning models in environments ranging from large datacenters to mobile devices. He will then discuss ways in which Google has applied this work to a variety of problems in Google's products, usually in close collaboration with other teams. This talk describes joint work with many people at Google.

Machine Learning Wars: Deep Learning vs Gradient Boosting [Webinar]Sefik Ilkin Serengil

��

This document discusses and compares gradient boosted machines (GBM) and deep learning. It notes that while deep learning can give powerful abilities, GBM may better explain models through decision trees. GBM works by combining many weak decision tree models to more accurately classify data, like moving a heavy rock bit by bit. The document cites GBM's success in Kaggle competitions, with it comprising the majority of winning solutions, sometimes combined with neural networks. In summary, the document presents GBM and deep learning, highlighting GBM's interpretability while also noting deep learning's strengths, and examples of GBM's competitive performance.

Machine Learning Wars: Deep Learning vs Gradient Boosting MachinesSefik Ilkin Serengil

��

This document compares and contrasts gradient boosting machines (GBM) and deep learning. It notes that while deep learning can perform complex tasks like facial recognition, its decisions are not understandable by humans. GBMs like gradient boosting decision trees address this issue by providing interpretable models with decision rules that can be traced back. The document also discusses popular GBM frameworks like XGBoost and LightGBM and notes that while GBMs are widely used in data science and won many Kaggle competitions, some experts still prefer neural networks.

More Related Content

Similar to Machine Learning meets Blockchain (20)

Hacktoberfest'24 _ GDG on Campus BU.pptxnilaygupta3003

��

Machine Learning in Cyber Security Domain BGA Cyber Security

��

building intelligent systems with large scale deep learningmustafa sarac

��

ButlerThomas Butler, CISSP, CEH, CHFI, CISA, CPA, CIA

��

DeepSeek AI Review: The AI Chatbot Shaping the Future of Conversational AIMasoomAkhtar2

��

Ethical Hacking, Its relevance and Its ProspectsRwik Kumar Dutta

��

9/30 Top 5 Deep Learning NVIDIA

��

How To Become an Ethical Hacker?Srashti Jain

��

Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014Austin Ogilvie

��

9/9/16 Top 5 Deep LearningNVIDIA

��

Citizen AI Engineer Program 2018 CAI 500 Fast Track AI Week1 RoadmapDr. Mohan K. Bavirisetty

��

HKUST Computer Science Festival 2013 - Seminar: Computer Science, Hacking and...Anthony Lai

��

Defining a Practical Path to Artificial Intelligence Roman Chanclor

��

The Programmable Internet of ThingsRich Miller

��

2015 Bio-IT Trends From the TrenchesChris Dagdigian

��

10/13 Top 5 Deep Learning StoriesNVIDIA

��

CompetitiveAdvantageThroughDeepLearning (white paper)Tarun Mehra

��

What does "monitoring" mean? (FOSDEM 2017)Brian Brazil

��

Notes on Simulation and GHDLDIlawar Singh

��

"Large-Scale Deep Learning for Building Intelligent Computer Systems," a Keyn...Edge AI and Vision Alliance

��

Hacktoberfest'24 _ GDG on Campus BU.pptxnilaygupta3003

��

Machine Learning in Cyber Security Domain BGA Cyber Security

��

building intelligent systems with large scale deep learningmustafa sarac

��

ButlerThomas Butler, CISSP, CEH, CHFI, CISA, CPA, CIA

��

DeepSeek AI Review: The AI Chatbot Shaping the Future of Conversational AIMasoomAkhtar2

��

Ethical Hacking, Its relevance and Its ProspectsRwik Kumar Dutta

��

9/30 Top 5 Deep Learning NVIDIA

��

How To Become an Ethical Hacker?Srashti Jain

��

Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014Austin Ogilvie

��

9/9/16 Top 5 Deep LearningNVIDIA

��

Citizen AI Engineer Program 2018 CAI 500 Fast Track AI Week1 RoadmapDr. Mohan K. Bavirisetty

��

HKUST Computer Science Festival 2013 - Seminar: Computer Science, Hacking and...Anthony Lai

��

Defining a Practical Path to Artificial Intelligence Roman Chanclor

��

The Programmable Internet of ThingsRich Miller

��

2015 Bio-IT Trends From the TrenchesChris Dagdigian

��

10/13 Top 5 Deep Learning StoriesNVIDIA

��

CompetitiveAdvantageThroughDeepLearning (white paper)Tarun Mehra

��

What does "monitoring" mean? (FOSDEM 2017)Brian Brazil

��

Notes on Simulation and GHDLDIlawar Singh

��

"Large-Scale Deep Learning for Building Intelligent Computer Systems," a Keyn...Edge AI and Vision Alliance

��

More from Sefik Ilkin Serengil (9)

Machine Learning Wars: Deep Learning vs Gradient Boosting [Webinar]Sefik Ilkin Serengil

��

Machine Learning Wars: Deep Learning vs Gradient Boosting MachinesSefik Ilkin Serengil

��

A Developer's Guide To Machine LearningSefik Ilkin Serengil

��

This document provides an overview of machine learning and perspectives from various experts: - It discusses different types of machine learning problems like classification, regression, and clustering and examples of algorithms used to solve each. - Experts offer views on neural networks, with one saying they are like a "swiss army knife" and can be used to solve many machine learning problems. - Other experts discuss the importance of linear algebra and matrix multiplication in machine learning models like neural networks. - One expert prefers neural networks and singular value decomposition for machine learning tasks.

Deep Learning, all well but how deep?Sefik Ilkin Serengil

��

Deep learning uses neural networks with many layers to learn representations of data with multiple levels of abstraction. It has been successfully used for tasks like image classification, speech recognition, and machine translation. Deep learning methods like convolutional neural networks have achieved human-level performance on some tasks, such as identifying objects in images. Transfer learning, where models pre-trained on large datasets are reused on other tasks, has also driven success in machine learning.

Deep Learning: Past, Present and FutureSefik Ilkin Serengil

��

Dive Deep Into Deep LearningSefik Ilkin Serengil

��

This document discusses deep learning and neural networks. It begins with an introduction of the author and their background. It then discusses topics like convolutional neural networks, their structure and applications to tasks like handwritten digit recognition and image classification. Examples of deep learning models that have achieved human-level performance on games, like DeepBlue defeating Kasparov at chess and DeepMind defeating Lee Sedol at Go, are provided. The document concludes with a question about guessing which of two pictures was painted by AI and thanks the audience for their attention.

Survival Guilde for Deep Learning AgeSefik Ilkin Serengil

��

Planning Workforce Management for Bank Operation Centers with Neural NetworksSefik Ilkin Serengil

��

Attacking Turkish Texts Encrypted by Homophonic CipherSefik Ilkin Serengil

��

Machine Learning Wars: Deep Learning vs Gradient Boosting [Webinar]Sefik Ilkin Serengil

��

Machine Learning Wars: Deep Learning vs Gradient Boosting MachinesSefik Ilkin Serengil

��

A Developer's Guide To Machine LearningSefik Ilkin Serengil

��

Deep Learning, all well but how deep?Sefik Ilkin Serengil

��

Deep Learning: Past, Present and FutureSefik Ilkin Serengil

��

Dive Deep Into Deep LearningSefik Ilkin Serengil

��

Survival Guilde for Deep Learning AgeSefik Ilkin Serengil

��

Planning Workforce Management for Bank Operation Centers with Neural NetworksSefik Ilkin Serengil

��

Attacking Turkish Texts Encrypted by Homophonic CipherSefik Ilkin Serengil

��

Recently uploaded (20)

Lecture 2-DATABASE MODELS lecture 2.pptxelvis24mutura

��

High-Paying Data Analytics Opportunities in Jaipur and Boost Your Career.pdfvinay salarite

��

PLAN_OF_WORK_PPT_BY_ROHIT_BHAIRAM_--2212020201003[1] new.pptxbhairamrohit948

��

Financial Ratios and CAMEL Presentation.pptPrinceAyangbesanOlam

��

data compression.ppt tree structure vectorvidhyaminnalveeran29

��

Capital market of Nigeria and its economic valuesezehnelson104

��

537116365-Domain-6-Presentation-New.pptxPorshaAbril1

��

Introduction to Microsoft Power BI is a business analytics serviceKongu Engineering College, Perundurai, Erode

��

Orange County Tableau User Group 2025 Late Q1 2025-03-23.pdfgemmajfrancisco

��

LITERATURE-MODEL.pptxdddddddddddddddddddddddddddddddddMaimai708843

��

STS-5 ----------------------------------leihlynnperegrin

��

Reason To Switch to DNNDNNs excel in handling huge volumes of data (e.g., ima...SrideviPcSenthilkuma

��

FinanceGPT Labs Whitepaper - Risks of Large Quantitative Models in Financial ...FinanceGPT Labs

��

Large Quantitative Models (LQMs) are a class of generative AI models designed for quantitative analysis in finance. This whitepaper explores the unique risks LQMs pose to financial markets, focusing on vulnerabilities to data poisoning attacks. These attacks can manipulate model outputs, leading to flawed economic forecasts and market instability. The whitepaper also addresses systemic risks like herding behavior and the potential for cascading failures due to the interconnectedness of financial institutions. Effective mitigation strategies, including robust data validation, adversarial training, real-time monitoring, and secure model development lifecycles, are discussed. The analysis emphasizes the need for proactive cybersecurity measures and regulatory frameworks to ensure the responsible and secure deployment of LQMs, maintaining the stability and integrity of financial markets.

SCIENCE-TECHNOLOGY-AND-SOCIETY.pptxhhgfffvTristanEvasco

��

ENG8-Q4-MOD2.pdfajxnjdabajbadjbiadbiwdhiwdhwdhiwdshekainahrosej

��

Heart Disease Prediction Using ML (274,332,167,546).pptxm23aid005

��

Basic Data Quality for beginner SAS coder.pptPhilipMason5

��

STS-PRELIM-2025.pptxtyyfddjugggfssghghihfTristanEvasco

��

Présentation d’idée idée pour faire un projettahatraval88

��

AI-Powered Contact Centre Virtual Assistant DSSrinivasan N

��

Lecture 2-DATABASE MODELS lecture 2.pptxelvis24mutura

��

High-Paying Data Analytics Opportunities in Jaipur and Boost Your Career.pdfvinay salarite

��

PLAN_OF_WORK_PPT_BY_ROHIT_BHAIRAM_--2212020201003[1] new.pptxbhairamrohit948

��

Financial Ratios and CAMEL Presentation.pptPrinceAyangbesanOlam

��

data compression.ppt tree structure vectorvidhyaminnalveeran29

��

Capital market of Nigeria and its economic valuesezehnelson104

��

537116365-Domain-6-Presentation-New.pptxPorshaAbril1

��

Introduction to Microsoft Power BI is a business analytics serviceKongu Engineering College, Perundurai, Erode

��

Orange County Tableau User Group 2025 Late Q1 2025-03-23.pdfgemmajfrancisco

��

LITERATURE-MODEL.pptxdddddddddddddddddddddddddddddddddMaimai708843

��

STS-5 ----------------------------------leihlynnperegrin

��

Reason To Switch to DNNDNNs excel in handling huge volumes of data (e.g., ima...SrideviPcSenthilkuma

��

FinanceGPT Labs Whitepaper - Risks of Large Quantitative Models in Financial ...FinanceGPT Labs

��

SCIENCE-TECHNOLOGY-AND-SOCIETY.pptxhhgfffvTristanEvasco

��

ENG8-Q4-MOD2.pdfajxnjdabajbadjbiadbiwdhiwdhwdhiwdshekainahrosej

��

Heart Disease Prediction Using ML (274,332,167,546).pptxm23aid005

��

Basic Data Quality for beginner SAS coder.pptPhilipMason5

��

STS-PRELIM-2025.pptxtyyfddjugggfssghghihfTristanEvasco

��

Présentation d’idée idée pour faire un projettahatraval88

��

AI-Powered Contact Centre Virtual Assistant DSSrinivasan N

��

Machine Learning meets Blockchain

1. What Happens If Machine Learning meets Blockchain Sefik Ilkin Serengil

2. sefiks.com serengil Sefik Ilkin Serengil Software Developer @ ‘10 CS graduate, MSc @ ’11

3. Data is the new oil (The Economist, 2017) 753B 739B72B782B 138B 2.2B Total: 2.4T $ France GDP: 2.5T $

5. Geoffrey Hinton Professor at University of Toronto, Fellow at Google Brain Inventor of Backpropagation - Learning Algorithm of Neural Networks Deep Learning is an algorithm which has no theoretical limitations of what it can learn; the more data you give and the more computational time you provide, the better it is

6. Andrew Ng Prof. at Stanford University Founder of Coursera, Chief Scientist at Baidu I have worked all my life in Machine Learning, and I’ve never seen one algorithm knock over benchmarks like Deep Learning

7. robertaguileradesign.com Fool your neural networks

9. Reported to have lost $136.5 billion in market cap

12. Adversarial Attacks

14. https://towardsdatascience.com/breaking-neural-networks-with-adversarial-attacks-f4290a9a45aa Classified as Speed Limit: 45 mph sign!

19. Peter Thiel Founder of Paypal Angel investor of Facebook Crypto is libertarian, AI is communist. https://www.youtube.com/watch?v=-iYpmOV60ks

20. Reid Hoffman Cofounder of Linkedin and Paypal Angel investor of Facebook Crypto is anarchy, AI is the rule of laws https://www.youtube.com/watch?v=-iYpmOV60ks

21. Blockchain is deterministic whereas ML is probabilistic

22. Blockchain has a decentralized architecture

23. ML has a centralized architecture

24. Blockchain transactions are transparent

25. Deep learning is a totally black box

26. Blockchain transactions are permanent

28. Deployment Time BusinessValue Model Model Remodel / retrain ML models are changing Model PositiveNegative

29. Secure & More Data Better Models Better Results Better Actions

30. sefiks.com serengil Thank you for your attention! Credit: Yasemin Dalkilic