Human language is broken, ambiguous, and outright bad. Bare with me.
Languages are a way we created to express ideas in a way that didn’t exist before.
Imagine a baby crying — you don’t know why he is crying, is he crying due to hunger, pain, not being able to see his mother. It’s confusing.
So to express ideas in a better way, we created human language.
But it was born at a time where humans were just planning stupid. It is wildly ambiguous — meaning a sentence can mean different things being read by different people. One could argue this makes it artistic and beautiful. You know what also does it do — it makes us inefficient. This is one of the reasons why computers are more efficient in certain tasks than humans. The computer language has proper Grammer and unambiguous. …
Hi, I am Divanshu Garg. I am an alumnus of IIIT Allahabad from 2015 batch. I obtained my bachelor's degree from the institution majoring in Information Technology. After graduating from college, I worked as a Software Engineer at Google Sydney for more than a year. Presently, I am working in a similar role at Google, Mountain View. I am enthusiastic about learning new technologies and building anything cool using them. I am particularly interested in machine learning, competitive programming and Android development.
The first day at college is always a memorable event for anyone and IIIT Allahabad made it even more so. I was fascinated by the beauty of the campus and the college infrastructure. The hostel quickly became an eternal part of my life. Right from the first day, I met so many great friends over there who made four years look short. Apart from batchmates, the seniors were also very helpful and provided guidance whenever I needed one. …
We’ve seen colour put into monochrome stills, silent movies being empowered with voice, and languages being translated with a single look. What paradigm-shifting innovations do you believe Deep Learning has in store for us all?
I hope that AI will help us to develop new medicinal techniques and green energy technologies.
Secondly, what do you think is underrated? These could be techniques that are not so well known or just ones that work well but aren’t popular/trendy.
Numerical stability. Very few people emphasize getting this right, but it has a large effect on the performance of deep learning algorithms. I decided to put a lot more about this in the lecture slides for the deep learning book than we were able to put in the book itself: http://www.deeplearningbook.org/slides/04_numerical.pdf …
When I was in high school (SFS Ettumanoor), we had this computer lab period. It was one of the most boring, no internet, sitting solving boring C++ questions.
The nerd in me stood among the few in this brave but short period of time, for we had found something that understood us more than humans — code. While others stood there crying in agony about the horrors of not even understanding the question.
The teacher had no idea, she had the networking on, so windows have this concept of shared files. …
A classifier is a machine learning model that is used to discriminate different objects based on certain features.
MS wants to connect with potential customers, our marketers and sellers.
people fill out online forms with fake names, gibberish, or even profanity.
A text can occur in a span or not so using a simple text detector will be bad.
But we can use the number of occurrence of each word in spam or not.
Hence we calculate probability of a word in spam
A new way to track your online and offline habits from your browser.
We made a new page chrome extension to motivate our small group when we open our browser. It works, so we wanted to share it with the world.
Introducing Karma ✔️
Discuss on Product Hunt
Click ✅ Done whenever you accomplished your target /habit.
Click on the habit name to toggle between habits.
List of most commonly asked questions on stack overflow about git.
by Mark Amery
Undo a commit and redo
git status, so you'll need to add them again before committing). …
Byzantine-fault tolerant state machine replication. Or blockchain, for short.
This is the first things you see when you open the Tendermint website. Odds are you already know what a blockchain is, It’s the technology which powers Bitcoin, Ethereum and other decentralized coins you heard of. The basic premise is that people can make applications in a decentralized way, and you can write in any programming language you prefer. Maybe you are confused so let’s back up a bit.
Tendermint is epic fast. Unlike others, you can send thousands of transactions in a second. …
Choosing a library is like choosing a wife, you have to stick with it and often changing one involves complicated process that is very expensive. So the best way to do this by analysis.
Now a days most analysis of libraries involve the features of the library. So we won’t get into that. We will go through this problem by example choosing a model server library for machine learning. A model server library allows you to easily make a server for your machine learning model.
This is a fairly obvious choice picking the one that gives you the most features. You look at your requirements and you see the library which gives you the minimal features you need. …