SlideShare a Scribd company logo

AI hype or reality

•Download as PPTX, PDF•
•
A
Awantik Das

The document provides an overview of machine learning and artificial intelligence concepts. It discusses: 1. The machine learning pipeline, including data collection, preprocessing, model training and validation, and deployment. Common machine learning algorithms like decision trees, neural networks, and clustering are also introduced. 2. How artificial intelligence has been adopted across different business domains to automate tasks, gain insights from data, and improve customer experiences. Some challenges to AI adoption are also outlined. 3. The impact of AI on society and the workplace. While AI is predicted to help humans solve problems, some people remain wary of technologies like home health diagnostics or AI-powered education. Responsible development of explainable AI is important.

Related slideshows

Moving from BI to AI : For decision makers
Moving from BI to AI : For decision makersMoving from BI to AI : For decision makers
Moving from BI to AI : For decision makers
 
Machine learning at scale - Webinar By zekeLabs
Machine learning at scale - Webinar By zekeLabsMachine learning at scale - Webinar By zekeLabs
Machine learning at scale - Webinar By zekeLabs
 
Machine Learning
Machine LearningMachine Learning
Machine Learning
 
1 of 76
zekeLabs AI - Hype vs Reality Learning made Simpler! www.zekeLabs.com
Modules ● Machine Learning Ecosystem ● AI In a Nutshell ● AI adoption across different domains ● Identify right tools ● Building AI Team
Module 1 ● Black box Introduction to Machine Learning ● Machine Learning Pipeline ● Adopting Machine Learning in your Product : Use cases Machine Learning Ecosystem
Black Box Introduction to ML
What is not Machine Learning ? ● Rule Based Approach ● Legacy Systems
Learning Algorithm What is Machine Learning ? ● Solve prediction problem ● Logic is learned from examples & not by rules Prediction Function or Trained Model Training Data Input Data Prediction
Types of Machine Learning Machine Learning ReinforcementUnsupervisedSupervised Task Driven Data Driven Environment Driven
Spam Mail Detection ● Input - Mail ● Output - Spam or Ham ● Supervised Machine Learning ● Binary Classification Problem
● Input - Sensor Data ● Output - Failure time ● Supervised Machine Learning ● Regression Problem Predicting Lift Failure
● Input - Accident details ● Output - Insurance amount ● Supervised Machine Learning ● Regression Problem Predicting Insurance Amount
● Input - Patient Synopsis (fever, temperature, BP, etc. ) ● Output - Diagnosis ● Supervised Machine Learning, ● Multi-class classification Problem Medical Diagnosis
Q. What is common between them ?
Market Segmentation ● Input - Customer Details ● Output - Clusters ● Unsupervised Machine Learning ● Clustering Problem
Robot playing Football ● Input - Player information, Rewards ● Output - Action to score ● Reinforcement Learning
Machine Learning Pipeline
Machine Learning Pipeline (MLP)
MLP - Business Understanding ● Business understanding includes clarity what you are trying to achieve. ● Machine learning is not possible with small data size ● Consolidating data pipeline to channelize continues flow of data. ● Web scraping, data lakes access, REST etc.
MLP - Data Wrangling ● Production data is never clean. ● It needs a major effort ( around 70% of total effort ) to make it ready for next stage ● Transforming & mapping data from raw format to another format ready for next stage
MLP - Data Visualization ● Visualization makes it easy to grasp difficult concepts ● Find useful pattern in the data ● Interactively drill down into charts for deeper details
Vectors - Fixed length array of numbers ● Text documents ● Image files ● CSV ● Audio ● Video ● Time Series data ● Many more ... MLP - Data Preprocessing Feature Extraction
MLP - Model Training Learning Algorithm Regression/Trees/SVM/Naiv e Bayes/Neural Networks/ Prediction Function or Trained Model
● Linear Regression ● Logistic Regression ● Naive Bayes ● Nearest Neighbors ● Decision Trees ● Ensemble Methods ● Clustering ● Support Vector Machines ● Neural Networks ● CNN ● RNN ● GAN MLP - Learning Algorithms
Prediction Prediction Function or Trained Model
MLP - Model Validation ● Training different learning method will give you different trained model. ● Also, each model have huge possibilities of configuration (hyper-parameters). ● Finding the best model among all possibilities & best configuration for it is done as a part of Model Validation. ● If results are not satisfactory, one has to go back in the chain & fix a few things
MLP - Deployment Trained Model Or Interface Model Consumers RESTful Interface
Adopting Machine Learning Real Stories
1. Customer Service Industry
1. Reduce manual effort of classifying reviews. 2.Channelizing data from Web server to Analytics Engine. 1. Getting data ready for visualization. 2. Historical data shows past trends. Visualization of trend Text needs to be tokenized & vectorized Different models were trained. Naive Bayes, SGD Classifier Choose the best model with best hyper- parameter Naive Bayes (MultinomialNB) was chosen & put in deployment 1. Customer Service Industry ● Manually labeled data is used for training model. ● Labels are target & review are feature data ● Batch training is supported by MultinomialNB allowing incremental learning ● Any mis-classification done by model will be labelled right & fed again
2. Fast Query Chatbots
2. Fast Query Chatbots 1. Reduce manual effort understanding the text query 2. Waiting for BI has a long turnaround time 3. We are trying to do this using chatbot 1. Getting data ready for visualization. 2. Historical data shows past trends Visualization of trend of text & sql Text cannot be used for ML Needs to be tokenized & vectorized Deep learning models with different layer configuration Choosing the best model with best hyper- parameter Model with best config was chosen & put in deployment ● Convert natural language query to SQL Query ● Model is trained with historical text (feature) & SQL (target) ● The generated SQL was executed & Output was subjected to visualization libraries ● Anybody without database & infra understanding can get visualization in seconds
3. Preventing System Failure
● Deep Learning - A specialization of Machine Learning ● ML vs DL vs AI ● AI Timeline ● What does AI consist of ? ● Where AI can be adopted in business ● Challenges in adopting AI Module 2 AI in a Nutshell
What is Deep Learning ? ● Specialized Learning Technique ● Rather than we choosing features for learning, this technique finds important feature derivatives. ● Objective is to learn best derived features for prediction. ● It mimics the way our brain learns ● Very useful for natural language, computer vision, audio, video etc.
Do you always need Deep Learning ? ● More data is required for Deep Learning ● More Compute Power ● Models less interpretable “Don’t kill a mosquito with a cannon ball” Don’t use Deep Learning if you don’t need to
ML vs DL vs AI. Timeline
What does AI consist of ?
Components of any AI product
Where AI got into in business?
Imp : Advice to executives about AI ● Everybody should embrace modern capability of AI, on other they should also think about business specific problems. Not every single tool that AI community can develop can suit them correctly. ● Biggest challenge is people change not technology change, biggest gap now is people who can map technology to business problem. ● Insourcing vs outsourcing. Building Team vs using enterprise solutions. ● AI will change everything in next few decades. Be a part of it.
Challenges - Data & Security ● Volume of data - Machine learning on smaller data is infeasible. ● Accessibility of data - Important data is not accessible & may be in encrypted format. info@zekeLabs.com | www.zekeLabs.com | +91
Compute, Storage & Network Power ● AI products needs data gathering from sensors, servers etc. ● Once gathered, data needs to be stored for further processing. ● Learning algorithms & data processing activities need lot of compute power.
Infrastructure for development ● Finding the best model is an iterative process. ● More experiments leads better model. ● Hyper-parameter Tuning ● Scaled infrastructure for developer is important.
Infrastructure for deployment ● Speedy Deployment ● Easy deployment ● Fluctuating Demand ● Need of Elastic infrastructure ● Cost optimization
Summary of Challenges
Cost optimization: ● Use Open Source alternatives ● Infrastructure optimization ● Don’t reinvent the wheel
AI hype or reality
● Will AI benefit human ? ● AI in human computer interaction ● Impact of AI on business ● Impact on workplace ● Impact on society Module 3 Impact of AI
AI benefit human - social, environmental ● Predicting diseases ● 60% People would prefer AI assistance over humans as financial advisors or tax preparers ● 71% people believe that AI will help humans solve complex problems and help live more enriched lives
AI Assistants ● Saves Time ● Calendar events reminder ● Helps get things done
Impact of AI on business
More
AI advisor & manager at workplace
Impact on Decision Makers ● Adoption of AI advisors
What can be outsourced to AI assistant?
Impact of artificial intelligence on society ● People are averse to the idea of availing annual health check-ups at home with a robotic smart kit (77%) or having chatbot assistant teachers in universities/ colleges that lower the cost of overall tuition (61%). ● Responsible AI ensures that its workings are aligned to ethical standards and social norms pertinent within its scope of operations. ● Explainable AI is responsible for building AI models with accountability and the ability to describe or depict why a certain decision was made by the algorithm.
● Programming Language ● Open source libraries ● Infrastructure Optimizations ● Other alternatives Module 4 Identify the right tools
Choose the Right Programming Language
Why Python makes life easy ? ● Easy to learn for ETL developers ● Integrates very well with other technologies ● Full-stack development - ○ Dashboard using bokeh, ○ Web application using django, ○ Machine learning models using scikit, ○ Scaling using PySpark
Choose appropriate Libraries - Statistical Modeling & Data Processing
Choose appropriate Libraries - Visualization
Choose appropriate Libraries - Machine Learning or Deep Learning
Infrastructure Optimization Monolithic or Serverless
Monolithic Infrastructure - Preallocated Infra Model Training ● Developers request access whenever required ● Might incur delay in peak working hours. ● Idle in non-working hours Model Interfacing ● Idle in non-peak hours. ● May fall short in spikes. ● Pay even if infra is not used
Serverless Infrastructure - Elastic Allocation Model Training ● No-preallocation ● Pay only for what you use ● Absolute no idle time for infra ● No wait time for developers Model Interfacing ● Allocate infra only when required ● Scales down during non-peak hours ● Improved customer experience even in peak hours
Serverless Infrastructure Solutions ● Open Function as a Service (OpenFaas) ● AWS Lambda ● Google Cloud Function ● Azure Function
Distributed Machine Learning using Spark ● Apache Spark is a distributed data processing framework. ● Many machine learning algorithms are implemented in Spark. ● Most of the API’s are same that of scikit-learn ● Scaled ETL & Machine Learning can be done using Spark
Other alternatives Google Cloud AI
● Adoption of AI ● Skills ● Hiring or upskilling ● Upskilling workforce Module 5 Build AI Team
Adoption Strategy Build Business Case Scale Efficiently Create Data Driven Culture
Skills
Talent Acquisition ● Upskill your current team ?
Upskilling workforce ● It’s possible to make use of the people who have delivered for you in the past.
Q & A
Repositories ● https://github.com/zekelabs/machine-learning-for-beginners ● https://github.com/zekelabs/tensorflow-tutorial/ ● Dog breed prediction - https://www.edyoda.com/resources/watch/54AEA4CDC35394F1183A9D D17AA47/ ● Python learning course - https://www.edyoda.com/resources/videolisting/98/ ● Learning Path - https://www.edyoda.com/program/data-scientist-program
Visit : www.zekeLabs.com for more details THANK YOU Let us know how can we help your organization to Upskill the employees to stay updated in the ever-evolving IT Industry. Get in touch: www.zekeLabs.com | +91-8095465880 | info@zekeLabs.com
Feedback QR code- DevelopU '19

More Related Content

AI hype or reality

  • 1. zekeLabs AI - Hype vs Reality Learning made Simpler! www.zekeLabs.com
  • 2. Modules ● Machine Learning Ecosystem ● AI In a Nutshell ● AI adoption across different domains ● Identify right tools ● Building AI Team
  • 3. Module 1 ● Black box Introduction to Machine Learning ● Machine Learning Pipeline ● Adopting Machine Learning in your Product : Use cases Machine Learning Ecosystem
  • 5. What is not Machine Learning ? ● Rule Based Approach ● Legacy Systems
  • 6. Learning Algorithm What is Machine Learning ? ● Solve prediction problem ● Logic is learned from examples & not by rules Prediction Function or Trained Model Training Data Input Data Prediction
  • 7. Types of Machine Learning Machine Learning ReinforcementUnsupervisedSupervised Task Driven Data Driven Environment Driven
  • 8. Spam Mail Detection ● Input - Mail ● Output - Spam or Ham ● Supervised Machine Learning ● Binary Classification Problem
  • 9. ● Input - Sensor Data ● Output - Failure time ● Supervised Machine Learning ● Regression Problem Predicting Lift Failure
  • 10. ● Input - Accident details ● Output - Insurance amount ● Supervised Machine Learning ● Regression Problem Predicting Insurance Amount
  • 11. ● Input - Patient Synopsis (fever, temperature, BP, etc. ) ● Output - Diagnosis ● Supervised Machine Learning, ● Multi-class classification Problem Medical Diagnosis
  • 12. Q. What is common between them ?
  • 13. Market Segmentation ● Input - Customer Details ● Output - Clusters ● Unsupervised Machine Learning ● Clustering Problem
  • 14. Robot playing Football ● Input - Player information, Rewards ● Output - Action to score ● Reinforcement Learning
  • 17. MLP - Business Understanding ● Business understanding includes clarity what you are trying to achieve. ● Machine learning is not possible with small data size ● Consolidating data pipeline to channelize continues flow of data. ● Web scraping, data lakes access, REST etc.
  • 18. MLP - Data Wrangling ● Production data is never clean. ● It needs a major effort ( around 70% of total effort ) to make it ready for next stage ● Transforming & mapping data from raw format to another format ready for next stage
  • 19. MLP - Data Visualization ● Visualization makes it easy to grasp difficult concepts ● Find useful pattern in the data ● Interactively drill down into charts for deeper details
  • 20. Vectors - Fixed length array of numbers ● Text documents ● Image files ● CSV ● Audio ● Video ● Time Series data ● Many more ... MLP - Data Preprocessing Feature Extraction
  • 21. MLP - Model Training Learning Algorithm Regression/Trees/SVM/Naiv e Bayes/Neural Networks/ Prediction Function or Trained Model
  • 22. ● Linear Regression ● Logistic Regression ● Naive Bayes ● Nearest Neighbors ● Decision Trees ● Ensemble Methods ● Clustering ● Support Vector Machines ● Neural Networks ● CNN ● RNN ● GAN MLP - Learning Algorithms
  • 24. MLP - Model Validation ● Training different learning method will give you different trained model. ● Also, each model have huge possibilities of configuration (hyper-parameters). ● Finding the best model among all possibilities & best configuration for it is done as a part of Model Validation. ● If results are not satisfactory, one has to go back in the chain & fix a few things
  • 25. MLP - Deployment Trained Model Or Interface Model Consumers RESTful Interface
  • 28. 1. Reduce manual effort of classifying reviews. 2.Channelizing data from Web server to Analytics Engine. 1. Getting data ready for visualization. 2. Historical data shows past trends. Visualization of trend Text needs to be tokenized & vectorized Different models were trained. Naive Bayes, SGD Classifier Choose the best model with best hyper- parameter Naive Bayes (MultinomialNB) was chosen & put in deployment 1. Customer Service Industry ● Manually labeled data is used for training model. ● Labels are target & review are feature data ● Batch training is supported by MultinomialNB allowing incremental learning ● Any mis-classification done by model will be labelled right & fed again
  • 29. 2. Fast Query Chatbots
  • 30. 2. Fast Query Chatbots 1. Reduce manual effort understanding the text query 2. Waiting for BI has a long turnaround time 3. We are trying to do this using chatbot 1. Getting data ready for visualization. 2. Historical data shows past trends Visualization of trend of text & sql Text cannot be used for ML Needs to be tokenized & vectorized Deep learning models with different layer configuration Choosing the best model with best hyper- parameter Model with best config was chosen & put in deployment ● Convert natural language query to SQL Query ● Model is trained with historical text (feature) & SQL (target) ● The generated SQL was executed & Output was subjected to visualization libraries ● Anybody without database & infra understanding can get visualization in seconds
  • 32. ● Deep Learning - A specialization of Machine Learning ● ML vs DL vs AI ● AI Timeline ● What does AI consist of ? ● Where AI can be adopted in business ● Challenges in adopting AI Module 2 AI in a Nutshell
  • 33. What is Deep Learning ? ● Specialized Learning Technique ● Rather than we choosing features for learning, this technique finds important feature derivatives. ● Objective is to learn best derived features for prediction. ● It mimics the way our brain learns ● Very useful for natural language, computer vision, audio, video etc.
  • 34. Do you always need Deep Learning ? ● More data is required for Deep Learning ● More Compute Power ● Models less interpretable “Don’t kill a mosquito with a cannon ball” Don’t use Deep Learning if you don’t need to
  • 35. ML vs DL vs AI. Timeline
  • 36. What does AI consist of ?
  • 38. Where AI got into in business?
  • 39. Imp : Advice to executives about AI ● Everybody should embrace modern capability of AI, on other they should also think about business specific problems. Not every single tool that AI community can develop can suit them correctly. ● Biggest challenge is people change not technology change, biggest gap now is people who can map technology to business problem. ● Insourcing vs outsourcing. Building Team vs using enterprise solutions. ● AI will change everything in next few decades. Be a part of it.
  • 40. Challenges - Data & Security ● Volume of data - Machine learning on smaller data is infeasible. ● Accessibility of data - Important data is not accessible & may be in encrypted format. info@zekeLabs.com | www.zekeLabs.com | +91
  • 41. Compute, Storage & Network Power ● AI products needs data gathering from sensors, servers etc. ● Once gathered, data needs to be stored for further processing. ● Learning algorithms & data processing activities need lot of compute power.
  • 42. Infrastructure for development ● Finding the best model is an iterative process. ● More experiments leads better model. ● Hyper-parameter Tuning ● Scaled infrastructure for developer is important.
  • 43. Infrastructure for deployment ● Speedy Deployment ● Easy deployment ● Fluctuating Demand ● Need of Elastic infrastructure ● Cost optimization
  • 45. Cost optimization: ● Use Open Source alternatives ● Infrastructure optimization ● Don’t reinvent the wheel
  • 47. ● Will AI benefit human ? ● AI in human computer interaction ● Impact of AI on business ● Impact on workplace ● Impact on society Module 3 Impact of AI
  • 48. AI benefit human - social, environmental ● Predicting diseases ● 60% People would prefer AI assistance over humans as financial advisors or tax preparers ● 71% people believe that AI will help humans solve complex problems and help live more enriched lives
  • 49. AI Assistants ● Saves Time ● Calendar events reminder ● Helps get things done
  • 50. Impact of AI on business
  • 51. More
  • 52. AI advisor & manager at workplace
  • 53. Impact on Decision Makers ● Adoption of AI advisors
  • 54. What can be outsourced to AI assistant?
  • 55. Impact of artificial intelligence on society ● People are averse to the idea of availing annual health check-ups at home with a robotic smart kit (77%) or having chatbot assistant teachers in universities/ colleges that lower the cost of overall tuition (61%). ● Responsible AI ensures that its workings are aligned to ethical standards and social norms pertinent within its scope of operations. ● Explainable AI is responsible for building AI models with accountability and the ability to describe or depict why a certain decision was made by the algorithm.
  • 56. ● Programming Language ● Open source libraries ● Infrastructure Optimizations ● Other alternatives Module 4 Identify the right tools
  • 58. Why Python makes life easy ? ● Easy to learn for ETL developers ● Integrates very well with other technologies ● Full-stack development - ○ Dashboard using bokeh, ○ Web application using django, ○ Machine learning models using scikit, ○ Scaling using PySpark
  • 59. Choose appropriate Libraries - Statistical Modeling & Data Processing
  • 61. Choose appropriate Libraries - Machine Learning or Deep Learning
  • 63. Monolithic Infrastructure - Preallocated Infra Model Training ● Developers request access whenever required ● Might incur delay in peak working hours. ● Idle in non-working hours Model Interfacing ● Idle in non-peak hours. ● May fall short in spikes. ● Pay even if infra is not used
  • 64. Serverless Infrastructure - Elastic Allocation Model Training ● No-preallocation ● Pay only for what you use ● Absolute no idle time for infra ● No wait time for developers Model Interfacing ● Allocate infra only when required ● Scales down during non-peak hours ● Improved customer experience even in peak hours
  • 65. Serverless Infrastructure Solutions ● Open Function as a Service (OpenFaas) ● AWS Lambda ● Google Cloud Function ● Azure Function
  • 66. Distributed Machine Learning using Spark ● Apache Spark is a distributed data processing framework. ● Many machine learning algorithms are implemented in Spark. ● Most of the API’s are same that of scikit-learn ● Scaled ETL & Machine Learning can be done using Spark
  • 68. ● Adoption of AI ● Skills ● Hiring or upskilling ● Upskilling workforce Module 5 Build AI Team
  • 69. Adoption Strategy Build Business Case Scale Efficiently Create Data Driven Culture
  • 71. Talent Acquisition ● Upskill your current team ?
  • 72. Upskilling workforce ● It’s possible to make use of the people who have delivered for you in the past.
  • 73. Q & A
  • 74. Repositories ● https://github.com/zekelabs/machine-learning-for-beginners ● https://github.com/zekelabs/tensorflow-tutorial/ ● Dog breed prediction - https://www.edyoda.com/resources/watch/54AEA4CDC35394F1183A9D D17AA47/ ● Python learning course - https://www.edyoda.com/resources/videolisting/98/ ● Learning Path - https://www.edyoda.com/program/data-scientist-program
  • 75. Visit : www.zekeLabs.com for more details THANK YOU Let us know how can we help your organization to Upskill the employees to stay updated in the ever-evolving IT Industry. Get in touch: www.zekeLabs.com | +91-8095465880 | info@zekeLabs.com
  • 76. Feedback QR code- DevelopU '19