Gamification is the use of game elements and game design techniques in non-game contexts.

More than one year ago, I found handy pages called Blinkist. It is a website worth checking if you don’t have much time to read whole books, but you want to improve yourself anyway. The pages provide a summary of hundreds of book in fields like productivity, management, self-improvements and many more nonfictional books. There is a free version where you will get one book per day, but you cannot choose which one. It is super if you have a problem with deciding … I bought a premium membership almost a year ago, and since then I have read nearly 100 books, so it was worth it - the cost is $6.67 per month if you pay the whole year in advance. You can also highlight parts of the books, and it is also the reason why I will be making these pages - to share my highlights from books which I have read there so far.

Marketing

I have been focusing on Action on Google (AoG), Dialogflow and Firebase for the last few weeks. Thanks to that I create a simple presentation and this blog post about the differences between Action on Google and Dialogflow. How to use them and how to connect them into one system based on device Google Home (or another Google Assistant enabled device).

Last weekend I was lucky to attend Machine Learning Prague 2019 for free thanks to the CzechAi. The whole conference takes two days and there was about 45 speaker, but the main part was meeting old friends and make the new one. In the end, I was surprised that there were so many great people in one place. I liked this conference mainly because of them.

This master’s thesis aims to find the best algorithm for semantic understanding of natural dialogs. The result can be used in a conversation AI. A part of the thesis is also a dataset, based on needs for a chatbot application for open-domain and which was based on real conversations. The thesis also identifies and examines sequential machine learning algorithms for intent and entity recognition. The result of the thesis is a detailed comparison of the selected algorithms regarding accuracy, memory requirements and computational complexity. Based on the results, a new model which joins intent and entity recognition together is created.

A PDF document with presentation of my master’s thesis.

Materials which I was using during preparation for the final exams:

ELMo (which stands for Embeddings from Language Model) is one of the new State-of-the-art technique to make word embeddings. I was trying to get the main principle and now I present a few highlights which were very useful for me to comprehend that.

These last few days I was attending WebExpo 2018 conference. On one side just because of the curiosity to see what is hype right now in Web. On the other side because I like networking and meeting new people. I was there like a volunteer - my usual way to get to conferences :D.

These last few days I was attending Human Level AI conference. On one side just because of the curiosity to see what is hype right now in an AGI (Artificial General Intelligence). On the other side because I like networking and meeting new people. Unfortunately the ticket price was not student-friendly so I became a volunteer to get on the conference for free and get at least few looks on some presentations. It was very interesting experience because almost all speakers had very good level of knowledge and were experts in theirs fields.

First attempt to summarize very broad topic of generative and discriminative models. I will be glad for every suggestion or discussion over this topic. I am sorry for math formulas but I wasn’t able to make it fully work yet. I will be trying to improve it over time.

I want to show you a way how to implement general custom class with defaultdict which you can nested together to create very useful structures.

Several thoughts and advices which I have been collecting since my journey to Taipei. There is no right answer, so you can just think about it by yourself and maybe change your behavior (I am also trying to do it as well). My plan is to make more posts like this.

Link to my presentation for Generating Sentences/Texts Using Deep Neural Network; 2. ML at Kx Systems Meetup

Because I was in that time in the Taipei and I am not Taiwanese, I was able to get free ticket to this big computer event. The event is divided into 4 different halls (3 close to Taipei 101 and one at Nangang).

Internet trends 2018 by Mary Meeker

The Entrepreneur’s Playbook: More than 100 Proven Strategies, Tips, and Techniques to Build a Radically Successful Business by Leonard C. Green.

The Startup festival 2017 (Prague) is a small event which is organized by students of University of Economics in Prague. The amount of shown startup wasn’t huge, but in this post I will focus on two workshops which were a part of additional program and I found interesting in some way.

The conference (WebExpo) is mainly about web, but it covers this topic in several views - design, business and development. I was happy to be part of the volunteer team, which help with conference in one day in exchange to free ticket to the other day. The first day I was helping with stage named “ABC Theatre” - in this hall was several interesting lectures which I was able to hear.

I just finished Neural Networks and Deep Learning course (first part) created by deeplearning.ai on the Coursera. I would like to write a little bit about my experience with it and also provide a little review.

When you use Google Allo or GMail on your mobile device, you have to see suggestion which the app provides when you receive a short message.

On the Internet is a lot of data (films/books/items/email). Almost nobody like to rummaging through the vast amount of things to find an interesting book or an unseen film, which I will like. For this purpose there is a huge development in recommendation systems. In this post I would like to provide a quick overview about this interesting topic.

There are two types of data - with and without labels. In this post I will focus on the unlabeled text, which should simulate human dialog or an email communication. There is a lot of stuff you can do with these texts, one of them is clustering.

I would like to write some posts about neural networks. This one is first and we will focus on the dimensionality. It is a basic principle in the neural network (to have the right dimension in matrix/vector multiplication).

My current research is about NLP with Czech text - I would like to use this interesting topic to be my diploma thesis, so I also decided to create this series to make accessible this interesting topic to everyone. In this series of post I would like to show some pitfalls which I have to encounter with. I will try to make as general as possible (it means that I won’t be focused on Czech text but on text in general).

Model situation: We need to decide true/false statement for some input (say image, which can be represented as 3 * array of color value (RGB values) and true would be if the number on the image is odd and false otherwise). What should we do or use??