Data Phoenix

2021-07-14 09:00:07 Reverse Engineering Generative Models from a Single Deepfake Image

Facebook AI in partnership with Michigan State University (MSU) presents a new method of detecting and attributing deepfakes. It relies on reverse engineering from a single AI-generated image to the generative model used to produce it.
https://bit.ly/3ko4xwS Open / Comment

2021-07-13 15:00:12 ​​TOP-5 tips for successful career in AI

Tip 1: Educational Requirements
According to teach.com, to start in the artificial intelligence field you’ll typically need a bachelor’s degree in IT, computer science, statistics, data science or a related field. For more advanced work, a master’s or PhD in one of these disciplines may be required.
Having strong STEM skills, including competency in statistics and mathematics, could also be beneficial in an AI engineer career.

Tip 2: Boost Up Your Skills
Beyond education requirements, the following are skills that are important for success:
Critical thinking and collaborative skills: A big part of your AI role will involve using data as a problem solving tool. This will require good communication and teamwork skills to effectively report and explain any insights found in data.
Analytical skills: Successful AI engineers are good with numbers. This requires an ability to think analytically, on the one hand, and to successfully communicate your thoughts and ideas to stakeholders, on the other.
Being business savvy: The ability to connect the dots between the practical world of business (use cases) and the specifics of the ML model (and data required in it) can be helpful to any AI engineer.

Tip 3: Learn Programming Languages
An AI professional should also demonstrate programming language proficiency in one or more of these common computer languages: Python, R, Java, C++

Tip 4: Keep Track of Tools and Frameworks
If you are interested in building up your career in artificial intelligence or you are searching for an artificial intelligence job, then you should know which framework or tool will make your code easy to implement.

Tip 5: Develop Your Own AI Project
You can develop a small project, or you can develop an ongoing project on GitHub. By developing an artificial intelligence project, you can check and test your own abilities.
Was it helpful? Let us know Open / Comment

2021-07-13 09:00:06 Effortless Distributed Training of Ultra-Wide GCNs

Graph independent subnetwork training (GIST) is a distributed training framework for large-scale graph convolutional networks (GCNs). It massively accelerates the training of GCNs for any architecture and can be used to enable training of large-scale models.
https://bit.ly/3k8CCRG Open / Comment

2021-07-12 15:06:26 Data Phoenix pinned «​​Happy Monday everyone! Here’s a quick reminder about our meetup this Wednesday. We at Data Phoenix, together with Autodoc and VITech, are excited to invite you to an offline meetup of Odesa’s Open Data Science community that’s going to take place July 14…» Open / Comment

2021-07-12 15:00:05 Catalyst.Neuro: A 3D Brain Segmentation Pipeline for MRI

In this article, you’ll learn about Catalyst.Neuro, an advanced brain segmentation pipeline, about its fundamental concepts implemented and different deep learning models to perform and complete brain segmentation tasks.
https://bit.ly/3ATuBpO Open / Comment

2021-07-12 09:20:10 ​​Happy Monday everyone! Here’s a quick reminder about our meetup this Wednesday.

We at Data Phoenix, together with Autodoc and VITech, are excited to invite you to an offline meetup of Odesa’s Open Data Science community that’s going to take place July 14, 6:30 PM — 9 PM. We’ll cover such topics as data management, object detection, and more. Most importantly, though, we’re going to network for real — that’s what we’ve been missing all these long quarantine months, right?

Because the number of seats is limited, we’re going to have an online session as well. The talks will be in Russian.

The event is free, but registration is required. So kindly register right away!
https://bit.ly/3hy0nRB Open / Comment

2021-07-11 11:00:21 ​​4 steps to get a successful AI project

1. Prioritize engineering over data science
As a rule of thumb, engineers can pick up data science skills faster than data scientists can pick up engineering skills. If you have any doubts about that, just work with any Python engineer with 5+ years of experience and passion for AI, rather than the PhD in data science having their first go at building business applications.
2. Go lean
It’s important to minimize risks early on. Structure your project around specific milestones. Make the team focused on launching «live» solutions (i.e. a pilot) in one to three months. When in production, decide whether further development will be worth it.
3. The algorithm doesn’t matter, data does
The algorithm is the least important part of your AI solution. Just choose an algorithm that works. Endlessly upgrading the algorithm is tempting, but it will probably not give you the results you expect. Focus on cleaning your data instead.
4. Communicate
Once the engineering team starts building, they have to make a lot of choices. The better they know your priorities, the more right decisions they can make.

Was it helpful? Let us know!
https://bit.ly/3r0iNxc Open / Comment

2021-07-10 11:00:05 ​​To get a better idea about how the industry works, sometimes it makes sense to watch a movie, just to get inspired by the power of talent and knowledge. Our team has prepared a few movies like that to demonstrate how machine learning can be depicted in a work of art.
Blade Runner 2049
One of the most impactful and in-depth storylines about smart machines. Blade Runner 2049 is a movie made for today’s machine learning age. In it, the replicants are “strong AI” — artificial general intelligence (AGI) that enables a machine to carry out any human task and even undertake advanced decisions on its own. Strong AI means replicants are stronger than humans in every respect and can even experience real emotions. However, the movie also deals with questions of soulless robots having consciousness and whether AI-powered intelligence means real consciousness.
https://bit.ly/36qO03d
Comment your favorite AI movie! Open / Comment

2021-07-09 14:10:05 ​​Are you tired of lockdowns? We for sure are!

We at Data Phoenix, together with Autodoc and VITech, are excited to invite you to an offline meetup of Odesa’s Open Data Science community that’s going to take place June 14, 6:30 PM — 9 PM. We’ll cover such topics as data management, object detection, and more. Most importantly, though, we’re going to network for real — that’s what we’ve been missing all these long quarantine months, right?

Because the number of seats is limited, we’re going to have an online session as well. The talks will be in Russian.
The event is free, but registration is required. So kindly register right away!
https://bit.ly/3hVZCkb
#meetup #DataScience #DataManagement #ObjectDetection Open / Comment

2021-07-09 10:00:11 ​​As a beginner, the first who you should know about is Geoffrey Hinton. He's a computer scientist and his biggest achievement is work on artificial neural networks. Hinton is famous as one of the most important figures in the deep learning community. Hinton was elected a Fellow of the Royal Society (FRS) in 1998. Additionally he was the first winner of the Rumelhart Prize in 2001. The fame of the AlexNet came when he in collaboration with his students Alex Krizhevsky and Ilya Sutskever designed for the ImageNet challenge 2012. This event was a breakthrough in the field of computer vision.
Geoffrey Hinton was one of the researchers who introduced the backpropagation algorithm and the first to use backpropagation for learning word embeddings. His other contributions to neural network research include Boltzmann machines, distributed representations, time-delay neural nets, mixtures of experts, variational learning and deep learning. His research group in Toronto made major breakthroughs in deep learning that revolutionized speech recognition and object classification.
https://bit.ly/3AKwl4r Open / Comment