I started the year with a few books/reading on machine learning. I took the class “Machine Learning” at Georgia Tech as being one of the required course to get the machine learning specialty on the master degree. The book in question was “Machine Learning” by Tom M. Mitchell. This book is tough to read. It is deep in the subject, and it is not written for a beginner in the field. The cost of
Get it Done
In parallel, I started the audiobook “Get it done” by Sam Bennett. It is a medium size reading of 6 hours 21 minutes. The book begins with the perspective that everyone is an artist and to transform fears into curiosity. There is the notion of performing something every day for 15 minutes. Procrastination is seen has three folds which are that you do not care or that the goal is a shadow goal which means that you have no joy executing. The second fold is that it might not be the right time to accomplish the task. Finally, procrastination might be a sign that the task build-up fears. I could resonate on the last point. The solution is to acknowledge the fear and to see the task as fun experimentations where failing is okay. In the end, having this attitude let you go through and most of the time success will be there. There is an allusion that most of the time small details appear more prominent than they are and to illustrate this idea the picture of looking back to something that was stressful might surprise you that what was causing the fear was more significant than it was in reality.
The book is a series of small tips to change the perspective when motivation become the culprit of moving forward. From basic idea like breaking down an idea into small chunks, to taking the time to ask “does this matter?” are proposition. There is the notion of “pure preference” which is to have only one target that might sound crazy but to keep on the list as something that we really would like to do. Comparing idea like if they were in competition one by one and assign a point to the winner until all of them are compared is another suggestion. Other suggestions are more funny at first. For example, to rename a task into something more funny which make it less boring when it is the time to execute it.
Here are a list of suggestions from the book:
- Create a “Could do” instead of “to do” list
- The way to present or talk about something might not be the good one
- It is important to develop a daily habit
- Breakdown ideas and organize them (shuffle)
- We have a left and right brain, we can keep the part of the brain that wants to do something by doing a repetitive movement (e.g., walking, taking a shower) to have the right part of the brain use its imagination to daydream.
- Keep track of the progress, write down all wins and learning (positive) because we have a tendency to have a bias toward negative
- Aim for a grade of ‘C’, not ‘A’ which is probably already more than the definition of a ‘C’ for other people.
- Ship early, once the ship sails it is easier to stop procrastinating and adding details. The example in the book is around “WordPress version 2”. It reduces the stress of delivering something perfect.
- Create a list of all that must be done with the cost of money, the time, the return on investment and determine what is really worth it.
- Instead of over-complicating something, ask people to find someone who already did it.
- Write a list of all the positive words people give to you. Refer to it.
- Always have deadlines. People have tendencies of executing on the last minute of them. People on time or early only move the deadline closer.
- Write down 30 ways to get what you want. Might be all possible but probably one of the items is good.
- Delegate what can be delegated
- Set clear boundaries between phone,
sms, email, work, relaxing time.
- Un-clutter your life. This section of the book is significant and take more time that I would expect. There
isfew good tricks but it was not the part that touched me much. One trick was interesting which was to pretend that someone needit, would you be willing to help that person? If naturally you would without hesitation, then it might be a clue that you can let go the object.
- Adding the word “sometime” in front of
a negativefeedback can help to digest the idea.
Finally, the book is good, but the last part was not interesting to me with trick about getting unclutter. However, the best advice I read was the following. “There is no shortage of success, there is no need to be jealous or to envy others”.
Influence – The psychology of persuasion
This book is from 2016 and
The book is packed with psychology studies and right in the first half hour I was hooked. The first advice is to stay simple. As much as possible. However, it does not mean to aim for simpler. You can do something very complex but you need to present it in a simple manner.
Giving a reason help tremendously when asking for something. The first chapter talk about some word that trigger some behavior. For example, the word “because” is very effective when desiring something.
The following notion is about “perceptual contrast” where our decisions are influenced by the former choice even if the choice was not accessible. For example, rating people while watching a movie with pretty people were causing people to rate lower if they were watching a movie with pretty people. Or, for
The principle of reciprocity is natural for every race, nation, time in history that when you do something nice for someone that the person will desire to do something nice. One tactic is to give something small (e.g. a flower) and expect the person to give later something of a bigger value or importance.
The principle of concession: if someone reduce what he wants in a negotiation, you will be prone to reduce also. This is why people starts with a higher bid and go down to what they really want.
The principle of consistency. Once someone is involved will suddenly be a strong defender of the idea even if a few seconds before taking the decision the person was on the edge. For example, if someone bet on a horse before a race, the person might hesitate but once the bet is settled, the person will be a fierce advocate.
We become what we do. E.g. POW that was threatened as collaborator started to behave like collaborator after a while.
People comply once they have one foot in the door more easily. E.g. Once someone accepts a small sign to be placed on their properties they had less resistance to change the sign into a big one later on.
Strategy of commitment. Once you commit to someone you do not want to break the promise hence would do more than initially planned to fulfill the engagement. E.g. Advertising a specific toy before Christmas with the plan to have kids to ask for the toy. The manufacturer does not produce a lot of toys to have parents buy another kind of toys. Then, after Christmas shows again the advertisement with abundant supply causing the kid to pressure the parent to fulfill the initial promise, hence getting the toy (additionally to the one from Christmas).
Writing something down increase commitment. Public commitment is stronger than having the decision only in your header. Human wants to look consistent.
The more it is hard to join, the more the feeling of belonging increase as well as solidarity among the people of the group. There is something around the increase of commitment. E.g. university performing hazing activity to join a student fraternity.
It is more efficient to have a very small reward for a big commitment. Motivation must be from inside the person to be strong in time. The person must feel a personal responsibility and not motivated by the reward or a consequence to behave. E.g. The kids who were saying that a teacher would be angry if they played with a particular toy. It was working until the user was substituted. The group of children who had to receive the consequence speech played with the toy above 70% while the other group who was explained why without reward or consequences played with less than 20%.
Social pressure has a big impact to motivate action. The book presents the example of a kid that does not want to swim until he plays with other kids who can do it and the example of how sect operate. The social proof is that in doubt people follow the herd. It is the concept of pluralist. However, unknown groups of people have less change to help because they are unsure of how the emergency is and people in the group do not want to look bad. The trick is to remove uncertainty. In case people are uncertain, they will copy others.
The halo effect is when one characteristic overshadows all other traits of someone. Physical trait (how beautiful is someone) is a common one but being kind is another. Similarity produces liking. A compliment works well, even if fake, even if known to be fake. People like more people that have something familiar. They will believe the person more if they can think alike or have a similar past.
The book slightly discusses the influence of seeing something much time that alter our judgment. E.g., picture flashing in sight. There is also guilt by association. E.g., a pretty girl next to a car will influence the first impression of the car from the pretty girl.
The idea of performing something kind create the feeling that the person is on the same side and hence will be willing to comply easily. E.g. if a waiter at a restaurant help you save some money on an appetizer, his tips might be bigger at the end or a future more expensive suggestion will be with a higher probability to be accepted.
The fear of losing something is bigger than the motivation to gain. However, hope is a bigger motivator than fear.
Scarcity is something that seems rare in appearance and the feeling worth more for people. It works with time as well.
The Hundred-Page Machine Learning
Another book I read is “The hundred-page machine learning book” by Andriy Burkov. I read this paperback book as a compliment for my first course. The book is short and does not cover every topic as I was expecting (for example KNN is 1 page long) but the first chapters were interesting for me. Mostly about the notation summary because some of the topics were far in my memory.