Blog
The Oldest AI Blog in the Czech Republic
We have been writing about artificial intelligence since 2017. 1000+ articles, thousands of pages of thoughts, experiments, and reflections. No sensationalism, no ads.
Tag Filter: Language Models × cancel
▸Browse by Topics
Topics
Showing 24 of 170 articles
Archive 2020FBIt's Here! GPT-3 is Coming! Fans of this page mostly already know that it is an AI model designed for text processing.
It's Here! GPT-3 is Coming! Fans of this page mostly already know that it is an AI model designed for text processing. There are 7 versions ranging from 125 million to 13 billion parameters! To top it off, OpenAI, which is behind this model,
Read
Archive 2020Have You Heard the Songs Sung by GPT-2?
OpenAI has introduced a new generative model that can create genre-specific music complete with lyrics! PS: It takes about 3 hours on a V100 GPU to generate 20 seconds of music...
Read
Archive 2020FBHave you heard the songs sung by GPT-2? OpanAI has introduced a new generative model...
Have you heard the songs sung by GPT-2? OpanAI has introduced a new generative model that can create genre-specific music complete with lyrics! PS: On a GPU V100, it takes about 3 hours to generate 20 seconds of music.
Read
Archive 2020Have you heard about the conversational agent, or chatbot, named Meena?
At the end of January, Google AI introduced it. It is a robust model with 2.6 billion parameters. The model's architecture is based on 13 decoding...
Read
Archive 2020FBHave you heard of the conversational agent, or chatbot, called Meena? In…
Have you heard of the conversational agent, or chatbot, called Meena? At the end of January, Google AI introduced it. It is a robust model with 2.6 billion parameters. The model's architecture is based on 13 decoding blocks of the Evolved Transformer.
Read
Archive 2019AI Surpasses Moore's Law
As you may know, Moore's Law states that approximately every 18 months (up to two years), the number of transistors on an integrated circuit doubles. A new report prepared by…
Read
Archive 2019FBAI Surpasses Moore's Law
AI Surpasses Moore's Law As you may know, Moore's Law states that approximately every 18 months (up to two years), the number of transistors on an integrated circuit doubles. A new report prepared by Stanford University and companies such as McKinsey &
Read
Archive 2019Another Course Completed
I have just completed the course ‘Learn BERT – the most powerful NLP algorithm by Google’. It is an advanced course that introduces…
Read
Archive 2019FBI have just completed the course "Learn BERT - most powerful NLP algorithm by Google".
I have just completed the course "Learn BERT - most powerful NLP algorithm by Google". It is an advanced course that introduces both theoretically and practically the use of the BERT model. Bertík is a cutting-edge algorithm from Google that...
Read
Archive 2019Artificial Intelligence Aids in Uncovering the Mysteries of Physics
It could one day help to reformulate quantum mechanics. Swiss physicist Rennato Renner and his team have proposed an algorithm that can distil vast amounts of information into…
Read
Archive 2019New Technological Trends from Gartner Symposium 2019
1. Hyperautomation is the new name for various automation techniques such as RPA, BPM, ML, DTO, NLP… 2. Multiexperience: People-Centric Smart Spaces. Technology is everywhere. The digital world…
Read
Archive 2019FBTechnological Trends from Gartner Symposium 2019:
Technological trends from Gartner Symposium 2019: 1. Hyperautomation is the new name for various automation techniques such as RPA, BPM, ML, DTO, NLP... 2. Multiexperience: People-Centric Smart Spaces. Technology is everywhere. The digital world surrounds us and
Read
Archive 2019Great news! OpenAI has finally released its largest XLGPT-2 model with 1.5 billion parameters and 48 layers.
I like that people find the outputs of GPT-2 to be very convincing. The 'credibility score' for the largest model is 6.91 out of 10. That's just slightly…
Read
Archive 2019FBGreat news! OpenAI has finally released its largest XLGPT-2 model with 1.5 billion…
Great news! OpenAI has finally released its largest XLGPT-2 model with 1.5 billion parameters and 48 layers. I like that people find the outputs of GPT-2 to be very convincing. The largest model has a 'credibility score' of 6.91 out of 10. That's just a little more than the outputs from the 774M model (6.72) and significantly more than the medium model 355M (6.07). The difference between the large and super large model is therefore relatively small (goodness, what a sentence:)). This was probably the last proverbial straw that led OpenAI to release the XL model.
Read
Archive 2019Enthusiasm and Concerns about Artificial Intelligence
Editors of the scientific editorial office of Czech Radio focused on artificial intelligence and the issues it brings. Here you will find sections on ethics, artificial intelligence in...
Read
Archive 2019FBLess than a month ago, the AI Salesforce team released a new text generation model, CTRL.
Less than a month ago, the AI Salesforce team released a new text generation model, CTRL. The algorithm works by allowing you to input a web link, and it attempts to generate an article that could be found there. This model is truly massive.
Read
Archive 2019The Latest Artificial Intelligence Model for Natural Language Processing – ALBERT!
Among the top artificial intelligence models for natural language processing (Bert, Robert, GPT-2, or Megatron), a new player ALBERT joined the ranks last week! ALBERT is brought to us by…
Read
Archive 2019FBAmong the Best Artificial Intelligence Models for Natural Language Processing (Bert, …
Among the best artificial intelligence models for natural language processing (Bert, Robert, GPT-2, or Megatron), another player has entered the field: ALBERT! ALBERT is brought to us by Google Research and the Toyota Technological Institute. What’s interesting is not just that the model delivers fantastic results on classic tasks like GLUE, RACE, or SQuAD, but also that it is smaller than its predecessors!
Read
Archive 2019Video: How to Utilise Multiple AI Agents?
Open AI presents a beautiful 3-minute video demonstrating how to utilise multiple AI agents for collaboration and tool usage. It’s a fascinating sight. Those adorable little figures managed to come up with solutions that human creators had no idea about :)
Read
Archive 2019Ladies and gentlemen, this week we are witnessing one groundbreaking step after another
First, Nvidia releases the colossal Megatron (the largest model in the world, five times bigger than the largest GPT-2), to which OpenAI responds by releasing its large model with...
Read
Archive 2019Nvidia Announces It Has Trained the Largest Language Model in the World, GPT-2 8B!
The model uses 8.3 billion parameters and is 24 times larger than BERT and 5 times larger than the previously largest GPT-2 from OpenAI. Nvidia employed parallelism that…
Read
Archive 2019Microsoft invests a billion dollars in OpenAI!
We write about OpenAI quite often. Not only were we amazed by their GPT-2 model, but we also noted how the famous Elon Musk helped establish it back in 2015, although he recently departed from the company.
Read
Archive 2019Open AI Comes with Another Interesting Project – MuseNet!
In this experiment with a deep neural network (using the same universal technology as GPT-2), instead of text, it attempts to predict musical notes. This is nothing new in AI,…
Read
Archive 2019Nietzsche Responded with AI to the Question: 'What is the Meaning of Human Life?'
Yesterday, I conducted a small experiment. I took my favourite model from OpenAI, called GPT-2. Imagine a vast artificial neural network the size of a bird's brain (with over 100 million parameters) trained on millions of interesting articles written by humans...
Read