Dataset bias strikes again! 😬

This is a great example of dataset bias! This ML method classifies skin lesions as being malignant or not.

Can you see the problem from the examples? 🤔

Let me shortly explain 👇

Machine Learning in the Real World 🧠 🤖

ML for real-world applications is much more than designing fancy networks and fine-tuning parameters.

In fact, you will spend most of your time curating a good dataset.

Let's go through the process together 👇

#RepostFriday

Collect Data 💽

We need to represent the real world as accurately as possible. If some situations are underrepresented we are introducing Sampling Bias.

Sampling Bias is nasty because we'll have high test accuracy, but our model will perform badly when deployed.

👇

Traffic Lights 🚦

Let's build a model to recognize traffic lights for a self-driving car. We need to collect data for different:

▪️ Lighting conditions
▪️ Weather conditions
▪️ Distances and viewpoints
▪️ Strange variants

And if we sample only 🚦 we won't detect 🚥 🤷‍♂️

👇

Data Cleaning 🧹

Now we need to clean all corrupted and irrelevant samples. We need to remove:

▪️ Overexposed or underexposed images
▪️ Images in irrelevant situations
▪️ Faulty images

Leaving them in the dataset will hurt our model's performance!

👇

Preprocess Data ⚙️

Most ML models like their data nicely normalized and properly scaled. Bad normalization can also lead to worse performance (I have a nice story for another time...)

▪️ Crop and resize all images
▪️ Normalize all values (usually 0 mean and 1 std. dev.)

👇

This is Karma. Karma is not a machine learning classifier 🐕‍🦺

Karma is a real dog trained to detect drugs. However, he would fail the simplest tests we apply in ML...

Let me take you through this story from the eyes of an ML engineer.

https://t.co/WAXRUlTvSI

Thread 🧵

Story TLDR 🔖

The story is about police dogs trained to sniff drugs. The problem is that the dogs often signal drugs even if there are none. Then innocent people land in jail for days.

The cops even joke about the “probable cause on four legs”.

Let's see why is that 👇

1. Sampling Bias 🤏

Drugs were found in 64% of the cars Karma identified, which was praised by the police as very good. In the end, most people don't carry drugs in their cars, so 64% seems solid.

There was a sampling problem though... 👇

The cars were not sampled at random! The police only did the sniff test if there was a serious suspicion that something is wrong.

The chance there are drugs in the car is much higher in this case!

2. Evaluation Metrics 🔍

The police referred to a 2014 study from Poland measuring the efficacy of sniffer dogs. The problem was that every test actually contained drugs!

This means there was no chance to measure false positives from the dogs! Only recall, not precision 🤦‍♂️

AI Job Interviews - another good example of bias in ML 🤦‍♂️

Two journalists tested some AI tools for assessing job candidates. Even when the candidate read a Wiki article in German instead of answering questions in English, the AI systems gave them good scores 🤷‍♂️

Let's unpack 👇

The Setup 🔬

The journalists created a fake job posting on two AI interview platforms. They specified the traits of the ideal candidate and provided the questions that need to be answered during the interview.

Then they started experimenting... 👇

The Positive Test ✅

One of them did a fake interview giving all the right answers and predictably got very high scores - 8.5 out of 9 👍

Then she tried something different... 👇

The Negative Test ❌

In a second interview, instead of answering the questions in English, she just read the article on psychometrics from the German Wikipedia 😁

One system gave her a score of 6 out of 9, while the other determined she is a 73% match for the job.

Oops... 👇

What happened? 🔍

Interestingly, one of the systems generated a transcript which was obviously meaningless.

This means that the machine learning model behind the tool likely captured nuances of the intonation of the speaker instead of the meaning of the actual words.

👇

Stories of Guruvayurappan
Shri Krishna Leela

This is a beautiful story of Guruvayurappan. Once there lived in Guruvayur a poor family. Kitta, his wife & kids were struggling for their daily needs. With only a coconut tree in house, he earned a living by selling coconuts.

Kitta was from a low caste family. He was a great devotee of Guruvayurappan. Everyday before going to work he always prayed infront of a picture of Guruvayurappan in his house. His only wish was to worship at Guruvayur & have annadanam in temple. But he was not allowed there

It was Utsavam time in Guruvayur. From far behind kitta and family worshipped Guruvayurappan. Annadanam was given daily and no one was left empty stomach. Only Kitta didn't receive. That night, it was only Kitta's family who was sleeping without food. After temple nada closed

Shri Krishna, dressed himself as a Brahmin went to temple Kitchen. He packed food and went directly to kitta's house. Reaching Kitta's house, Shri Krishna told he was sent by a cook of Guruvayur temple and gave them the food he packed. Kitta and family was very happy seeing food.

When Kitta finished his food, Brahmin asked it will be better if he get a tender coconut( karikku)to drink. Kitta climbed the tree and gave all coconuts to him. After having 1 tender coconut, the Brahmin asked to bring remaining coconuts to Guruvayur temple the next day.

Again sharing bank nifty intraday weekly straddle. How to play by selling straddle.
Mostly will work on Monday to Wednesday. Sometimes on Friday.
Why to pay your money to learn this if u can get it here.
U need only practice.
Let’s go.

Say bank nifty future is at 32000.
Sell both 32000 call and put.
Considering current volatile market lest say premium on Monday morning is 350 on both sides.
Sold with 700 points combine premium.
If bank nifty will trade in narrow range say +\- 100 hold it.
If moving 200 points

then exit previous sold straddle. Sell new one. Let’s say BN moved +200
Now sell 32200. If bank nifty will move here within range then u will get profit through theta decay. Follow same procedure for next 150-200 pts move.

This is only for single straddle sell position.

Now in second case sell three straddle. 31900/32000/32100
In case of 100 points move no need to change.
If 200 + then future is 32200.
Exit 31900 and sell 31200 straddle.
If moving another 100 then exit 32000 and sell 32300. Continue same with 100 points move.
If no move then

it will consolidate between 32000-32200 or 32100 to 32300 and u will get theta decay.

This is for normal move. Sudden big move may come anytime. So keep SL in system.
If u r selling at 350-350 then keep SL in system 475 for both legs.
Practice it. No need to attend webinar.

This is Kuinini ‘Nini’ Manumua, the woman who was ultimately displaced by inclusion of Laurel Hubbard.

She’s 21, and it would have been her first Olympics.

@AchuthArora @aldothetrainer @Artofphysique_ @buzz_chronicles @rattibha Gotcha, boss!

Your tweet has been saved.

Please sign up with savetobookmarks .com in order to view and manage your bookmarks.

@buzz_chronicles @AchuthArora @Artofphysique_ @rattibha Ahoy!

Now your tweet was bookmarked.

To access and manage your bookmarks, please register with savetobookmarks .com.