This is lecture 3 in the series on Wasserstein #GAN. In this lecture, basic understanding of Wasserstein Generative

Adversarial Network (WGAN) is discussed

videos


✴️ @AI_Python_EN

✴️ @AI_Python_EN

#Mathematics for #artificialintelligence

https://rubikscode.net/2019/04/29/mathematics-for-artificial-intelligence-linear-algebra/

Ironically, Yuval Noah Harari's equation of B X C X D= HH, where B=biological knowledge, C=computer power, D=data, HH=human hacking in days after the 1st report of direct #brain activity to speech.

Fei Fei Li to YNH : "Okay, can I be specific? First of all the birth of AI is AI scientists talking to biologists, specifically neuroscientists, right. The birth of AI is very much inspired by what the brain does. Fast forward to 60 years later, today's AI is making great improvements in healthcare. There's a lot of data from our physiology and pathology being collected and using machine learning to help us. But I feel like you're talking about something else."

https://www.wired.com/story/will-artificial-intelligence-enhance-hack-humanity/

✴️ @AI_Python_EN

What are histograms?
• Histograms are collected counts of data organized into a set of predefined bins
• When we say data we are not restricting it to be intensity values (as we saw in the previous Tutorial). The data
collected can be whatever feature you find useful to describe your image.
• Let’s see an example. Imagine that a Matrix contains information of an image (i.e. intensity in the range 0-255):

What happens if we want to count this data in an organized way? Since we know that the range of information
value for this case is 256 values, we can segment our range in subparts (called bins) like:
[0; 255] = [0; 15] ∪ [16; 31] ∪ :::: ∪ [240; 255]
range = bin1 ∪ bin2 ∪ :::: ∪ binn=15
and we can keep count of the number of pixels that fall in the range of each bini. Applying this to the example
above we get the image below ( axis x represents the bins and axis y the number of pixels in each of them).


This was just a simple example of how an histogram works and why it is useful. An histogram can keep count
not only of color intensities, but of whatever image features that we want to measure (i.e. gradients, directions,
etc).
• Let’s identify some parts of the histogram:
1. dims: The number of parameters you want to collect data of. In our example, dims = 1 because we are
only counting the intensity values of each pixel (in a greyscale image).
2. bins: It is the number of subdivisions in each dim. In our example, bins = 16
3. range: The limits for the values to be measured. In this case: range = [0,255]
• What if you want to count two features? In this case your resulting histogram would be a 3D plot (in which x
and y would be binx and biny for each feature and z would be the number of counts for each combination of
(binx; biny). The same would apply for more features (of course it gets trickier).
The OpenCV Tutorials, Release 2.4.13.0

✴️ @AI_Python_EN

Activation Atlases: a new technique for visualizing what interactions between neurons can represent

By Google and OpenAI.

Blog: https://blog.openai.com/introducing-activation-atlases/
Paper: https://distill.pub/2019/activation-atlas
Code: https://github.com/tensorflow/lucid/…
Demo: https://distill.pub/2019/activation-atlas/app.html

#artificialintelligence #deeplearning #machinelearning #neuralnetworks

✴️ @AI_Python_EN

free courses

🔸 Machine Learning (University of Washington)
🔸 Machine Learning (University of Wisconsin-Madison)
🔸 Algorithms (in journalism) (Columbia University )
🔸 Practical Deep Learning (Yandex Data School)
🔸 Big Data in 30 Hours (Krakow Technical University )
🔸 Deep Reinforcement Learning Bootcamp (UC Berkeley(& others))
🔸 Introduction to Artificial intelligence
(University of Washington)
🔸 Brains, Minds and Machines Summer Course (MIT)
🔸 Design and Analysis of Algorithms
(MIT)
🔸 Natural Language Processing
(University of Washington)

🌎 link

#MachineLearning #DataScience #Course #DeepLearning #BigData #AI

✴️ @AI_Python_EN

Real numbers, data science and chaos: How to fit any dataset with a single parameter

Paper:
https://arxiv.org/abs/1904.12320

Code:
https://github.com/Ranlot/single-parameter-fit/

#artificialintelligence #datascience #dataset #machinelearning

✴️ @AI_Python_EN

Artificial Intelligence and Games by Georgios N. Yannakakis

🌎 Book

#artificialintelligence

✴️ @AI_Python_EN

What will be the #programming_language for machine learning in the next few years?
Now it's #Python, but I would rather wish for better language because:

1- Python it's too slow. it's slower than JS!! to work around this problem we either use #C++ in the backend or use other technologies to Python faster such as Cython PyPy.

2- Doesn't support smooth and real Multiprocessing.
So we need a new programming language that has:

1- Nice syntax and easy to learn

2- Fast enough (at least not slower than #JS) without using complex tools.

3- Numerical computing and machine learning ecosystem

4- (Optional) integrated linear algebra operations such as adding two vectors.
I feel it could be either Julia or Swift
#machinelearning

✴️ @AI_Python_EN

Deep Learning Drizzle - Github by Marimuthu K.
https://lnkd.in/f6TANzj


Contents
=======

✅Deep Learning (Deep Neural Networks)

✅Machine Learning Fundamentals

✅Optimization for Machine Learning

✅General Machine Learning

✅Reinforcement Learning

✅Probabilistic Graphical Models

✅Natural Language Processing

✅Automatic Speech Recognition

✅Modern Computer Vision

✅Boot Camps or Summer Schools

✅Bird's Eye view of Artificial (General) Intelligence

#deeplearning #machinelearning #datascience

✴️ @AI_Python_EN

1. Data Science Implementation
https://lnkd.in/fMHtxYP

2. Machine Learning Implementation
https://lnkd.in/f5aUbBM

3. Discovery Analytics Cheatsheet
https://lnkd.in/f396Dqg

4. Machine Learning Key Terminology
https://lnkd.in/fCihY9W

5. SQL Cheatsheet
https://lnkd.in/fKyki2j

6. Machine Learning Cheatsheet
https://lnkd.in/fezaQme

7. Docker Chetasheet
https://lnkd.in/ffMrZXj

8. Tutorial Biglist
https://lnkd.in/fyFxQsM

9. Git Cheatsheet
https://lnkd.in/fWSHH_x

10. Self Driving Car
https://lnkd.in/fxMNBEh

11. Heathcare
https://lnkd.in/fn-yWSD


12. Data Science Cheatsheet
https://lnkd.in/fJgruHJ

✴️ @AI_Python_EN

A few years ago "machine learning" joined the ranks of terms that have lost much of their original meaning and largely become sales jargon. "Insight", which often simply means number, is another example.

Machine learning is sometimes used synonymously with Artificial Neural Networks but can also mean good old fashioned regression, K-means clustering, principal components analysis...or computer program.

If you'd like to learn more about what machine learning really is and how it is used in practice, here are some popular data science books I can recommend:

- Data Mining (Whitten et al.)
- Applied Predictive Modeling (Kuhn and Johnson)
- An Introduction to Statistical Learning (James et al.)
- Elements of Statistical Learning (Hastie et al.)
- Data Mining: The Textbook (Aggarwal)
- Machine Learning: A Probabilistic Perspective (Murphy)
- Pattern Recognition and Machine Learning (Bishop)

These books are somewhat technical but even a glance at the TOC and a peek insight might be helpful.

I am not suggesting machine learning is BS - I use some form of it nearly every day. The hype surrounding the term, however, confuses a lot of people and some may be shelling out big bucks for something we used to just call software.

I'd better not get started on "AI"... :-)
#machinelearning #artificialintelligence

✴️ @AI_Python_EN

Here’s tips to build business portfolio

✅ Step 1. Understand Data Science Implementation
https://lnkd.in/fMHtxYP

✅ Step 2. Understand what business want
https://lnkd.in/f396Dqg

✅ Step 3. Know Machine Learning Key Terminology
https://lnkd.in/fCihY9W

✅ Step 4. Know Business Implementation of Data Science
https://lnkd.in/f5aUbBM

If you have problem to get datasets in implement those 4, you can try a list of 10 GREAT
#SelfStarting #DataScience Projects to work on:

➖BEGINNER➖

1. Pokemon - Weedle's Cave 🐛
Python - https://lnkd.in/gcKWWQ2

2. Titanic ML 🚢
Python - https://lnkd.in/gafie9m

3. Housing Prices Prediction 🏡
Python - https://lnkd.in/gX2FSDk


➖INTERMEDIATE➖

4. Instacart Market Basket Analysis 🛒
Python - https://lnkd.in/gkNaXqH

5. Quora Question Pairs 👥
Project :https://lnkd.in/f3HQZsT
Tutorial (Python)- https://lnkd.in/fEzf-Xp

6. Human Resource Analytics 🕴🏻
Python - https://lnkd.in/gVUPfWm


➖ADVANCED➖

7. Analyzing Soccer Player Faces ⚽️
Python - https://lnkd.in/gUys_TS

8. Recruit Restaurant Visitor Forecasting 🍱
Python - https://lnkd.in/gjQvf74

9. TensorFlow Speech Recognition 🗣
Python - https://lnkd.in/g8SSPfW


➖MASTERY➖

10. More complete guide from Analytics Vidhya
https://lnkd.in/g_QjzGe.

#ml #forecasting #analytics

✴️ @AI_Python_EN

Data science is an ever-evolving field. As data scientists, we need to have our finger on the pulse of the latest algorithms and frameworks coming up in the community.

So, if you’re a:
Data science enthusiast
Machine learning practitioner
Data science manager
Deep learning expert

or any mix of the above, this article is for you.

more to read : https://bit.ly/2Jb2JoB

#machinelearning #datascience #deeplearning #deeplearning

✴️ @AI_Python_EN

Neural nets had been batting SVM for supremacy in data mining (aka data science) since the latter's rebirth thanks to the discovery of the kernel trick in the mid-90s. SVM seemed to be winning but then deep learning enjoyed its own resurrection and computing power began to make a wider variety of ANN architectures feasible. When I was a "kid", neural nets were mainly single layer MLP or Kohonen clustering (SOM). For what little it's worth, I see reinforcement learning as the key to true AI, though other forms of learning will continue to play a role.

#machinelearning #artificialintelligence

✴️ @AI_Python_EN

Open Source Resource for Newbies to Learn Algorithms and Implement them in any Programming Language.

*** The Algortithm ***

Github Link - https://lnkd.in/edw2vHj

#pythonprogramming #python #java #scala #c #cplusplus #csharp

✴️ @AI_Python_EN

check this collection of awesome deep Learning resources

https://github.com/frontbench-open-source/Data-Science-Free

#deeplearning

✴️ @AI_Python_EN

If you want to get into AI, what should you do?


Allie K. Miller give sharp and clear answer: "it completely depends"

This is decision tree where to start.

Follow the questions to find your first step to get a little closer to your AI goals, whatever they may be.

#artificialintelligence #AI #machinelearning

✴️ @AI_Python_EN

Some random thoughts on marketing mix modeling...

The basic purpose is to estimate the impact of a brand's marketing on its sales. It is also known as market response modeling and I've also heard the term sales response modeling used.

The simplest approach is to compare brands cross-sectionally and see how they differ in terms of sales and marketing activity. This is ill-advised because it does not account for how the brands' marketing and sales have changed over time.

Another popular way is to analyze (for example) four years of weekly sales and marketing data with multiple regression. The problem here is that the data are treated as cross-sectional whereas they are time-series data.

There are many kinds of time-series analysis which, properly conducted, can give us better estimates of the effects of marketing on sales. There are statistical considerations but one reason is that lagged effects can be estimated.

The problem with this approach is that marketing activity is assumed to be exogenous; in reality, marketing budgets and marketing plans are usually influenced by a brand's own sales and share, competitors' sales and share, competitors' marketing activity and by other exogenous factors such as the economy.

Whatever method we use, it must account for endogeneity.

Note: that these core issues apply far beyond mix modeling and even econometrics, where many of the methods we use in mix modeling originated. (My post was actually sparked by a paper on how to adjust for measurement artifacts in surface temperature data, e.g., land use patterns near weather stations.)

✴️ @AI_Python_EN

Jure Leskovec, Professor at Stanford


Deep Generative Models for Graphs: Methods & Applications.
Slides from my talk at #ICLR2019 workshop on Representation Learning on Graphs and Manifolds.
https://lnkd.in/eTHRDsM

#deeplearning

✴️ @AI_Python_EN