.png){kind=logo}
Welcome to the AI-ML Workshop
Decrypting the AI-ML hype and how artificial intelligence is reshaping the World!
Series of workshop hosted by TechConverge, SVNIT.
What is Machine Learning?
* * *
Definitions of ML
-
Field of study that gives computers the ability to learn without being explicitly programmed
Definitions of ML
-
"Field of study that gives computers the ability to learn without being explicitly programmed."
Arthur Samuel -
"A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E."
Tom Mitchell
Why Machine Learning is important?
-
Automation
It has allowed companies to transform processes that were previously only possible for humans to perform—think responding to customer service calls, bookkeeping, and reviewing resumes for everyday businesses.
-
Handling large problems
-
Exponential growth of ----
Why Machine Learning is important?
-
Automation
-
Handling large problems
Think of the image detection for self-driving cars, predicting natural disaster locations and timelines, and understanding the potential interaction of drugs with medical conditions before clinical trials.
-
Exponential growth of ----
Q. How all has this even become possible?
Ans: Data
Why Machine Learning is important?
-
Automation
-
Handling large problems
-
Exponential growth of DATA
This era of Information therefore has enhanced engagement in this information exchange.All this activity is resulting in tons of data being pumped out — Big Data
When Big Data meets Machine Learning the models flourish!!
-
Data is cheap nowadays at least relatively; Knowledge is harder to come by.
-
From data we are able to derive value from large, heterogeneous and disparate sources of data at lightning speed and scale.
-
ML helps to turn data into knowledge.
-
Marriage of Data and ML
-
Medical Diagnostics
The amount of knowledge available about a certain task might be too large for explicitly encoding by Humans
-
Data’s importance in ML
-
Automation
Marriage of Data and ML
-
Medical Diagnostics
-
Data’s importance in ML
Machine learning data analysis uses algorithms to continuously improve itself over time, but quality data is necessary for these models to operate efficiently.
-
Automation
Marriage of Data and ML
-
Medical Diagnostics
-
Data’s importance in ML
-
Automation
The task that can be completed with a datadefined pattern can be automated with the help of ML
Outcome of this Marriage
-
Virtual Assistant
-
Autocorrection
-
Autocompletion
Spoiler Alerts
Machine learning is not the answer to all problems.
-
Limitation 1 — Ethics:The idea of trusting data and algorithms more than our own judgment has its pros and cons.
-
Limitation 2 — Data:Manifest itself in two ways: lack of data, and lack of good data.
-
Limitation 3 —Interpretability:If you cannot convince your client that you understand how the algorithm came to the decision it did, how likely are they to trust you and your expertise?
Confusing data
Type of Machine Learning
This is how machine learn Supervised!!?
-
Train
-
Test
-
Validation
.png)
This is how machine learn Supervised!!? (Formal way)
Given (X, Y), find the mapping between Y → X
We wanted to predict price of a house, give some features of a house.
edited.png)
This is how machine learn Supervised!!? (Formal way)
Given (X, Y), find the mapping between Y → X
We wanted to predict price of a house, give some features of a house.
With the help of this line we can predict Y given X
Here Y is a label
Regression vs. Classification
edited.png)
Regression
Prediction output belongs to the set of Real No.
Prediction output is continuous in nature.
Regression vs. Classification
Regression vs. Classification
Classification
Prediction output belongs to the set of discrete No. like here we have two classes. output will be either 1 or 2
Prediction output is a class.
Regression vs. Classification
The Unsupervised way
The Unsupervised way
Given an Unlabeled data , Algorithm should make groups out of them.
For ex: Social media recommendation , Netflix recommendation
The Unsupervised way
The Cocktail Party Problem
Independent component analysis (ICA) Algorithm is used for it
Reinforcement Learning
Reinforcement Learning
Model which make intelligenent decisions
for each decision Reward or penalty is assigned to agent
AI is trained to maximise the reward
Conclusion
Where should I start?
Prerequisites
These are some prerequisites from our side.
- Basic Python: basic knowledge of python as well as list,dict,loops and if-else block
- Python Libraries: numpy,pandas,matplotlib
- Linear Algebra:understanding of matrix from 12th std.
- Basic Calculus: understanding of differenciation from 12th std.
- Probability and Statistics: bayesian theorem , mean,std. deviation,