Machine Learning Hackathon in Gdańsk

ML Gdańsk together with HACKER:SPACE Trójmiasto has organized Machine Learning Hackathon in Gdańsk. This event has been sponsored by PredictX. It was very energetic and collaborative meeting:
  • 2 coaches,
  • 20 attendees,
  • 2 different problems (classification & regression),
  • 6 hours of collaboration and networking.
Event Log:
Piotr + Bartosz – general info about the sponsor – PredictX.
Robert – introduction to classification, PhysioNet EKG challenge.
Marcin – introduction to time series analysis.
– Ramp-up with data & problems.
– Pizzas + hamburgers + salads.
– Working on problems.
– Presentation of results.
All attendees has decided on classification problem:
Few solutions have accuracy in the range of 60-75%.
However, it is difficult to compare with the official final result of the challenge (83%), because the participants tested it on the basis of a non-standard train / test split.
Event announcements:
Hackathon summary on Piotr W. blog (in Polish):
Information on the event’s sponsor:
  • PredictX delivers big data, machine learning, and decision automation services to data-reliant industries, in particular financial services, procurement, retail, travel and healthcare. Using embedded intelligence and data visualisation technology, PredictX continuously drive cost savings, operational efficiencies and revenue growth. PredictX supports clients in capturing, cleaning and compiling their data from multiple sources from within and outside the business. More: https://www.predictx.com/

40½ Meeting – The Deep Learning Revolution: From Technology to Impact

10 May (Thursday) – special meeting, due to the special guest – Chris Rowen – who visited Gdańsk and presented passionate talk about the machine learning revolution, accelerating startups, and especially about his new startup – BabbleLabs.

Chris is a well-known Silicon Valley entrepreneur and technologist, now serving as co-founder and CEO of BabbleLabs. He has led Cognite Ventures, a specialized analysis, investment and advisory company for deep learning start-ups. He has served as CTO for Cadence’s IP Group, where he and his team develop new processor and memory for advanced applications in mobile, automotive, infrastructure, deep learning and IoT systems. Chris joined Cadence after its acquisition of Tensilica, the company he founded to develop extensible processors. He led Tensilica as CEO and later, CTO, to become one of the most prolific embedded processor architectures.

Before founding Tensilica in 1997, he was VP and GM of the Design Reuse Group at Synopsys. Chris also was a pioneer in developing RISC architecture and helped found MIPS Computer Systems, where he was VP of Microprocessor Development. He holds an MSEE and PhD in electrical engineering from Stanford and a BA in physics from Harvard. He holds more than 40 US and international patents. He was named an IEEE Fellow in 2015 for his work in development of microprocessor technology.

20th meeting – SVM

The 20th meeting of our group took place on 22th of May.

The topic of the meeting was Support Vector Machines. It was thoroughly described by Krzysztof Czarnowski. The presentation was followed by interesting discussion about some mathematical details of those algorithms.

Next meeting will be held on 5th of June.

 

19th meeting – decision trees and random forest

During our 19th meeting on 8th of May, Marcin Zadroga talked about decision trees and random forest methods.

Marcin described the concepts and intuitions which are basis for those popular algorithms. Also the methods like bagging and boosting were discussed.

Next meeting of our group will be held on 22nd of May. Join us!

18th meeting – logic programming in Scheme / ECG signal analysis

On 24th of April 2017 we had 19th meeting of our group.

Two topics were presented. First presenter was Maciej Godek, who talked about logic programming in Scheme. He also showed us some programs he created – neural networks and genetic algorithm implementations. Additionally, Maciej presented the framework for creating virtual robots.

Second presentation was delivered by Kuba Domaszewicz. The topic of this talk was ECG signal analysis. Methods described by Kuba are used in aidlab device (health tracking wearable assistant).

Next meeting will be held on 8th of May. Feel free to join us!

 

17th meeting – Alpha Go / image recognition in security domain

Our 17th meeting was on 3rd of April 2017.

This time we had opportunity to listen to two speakers. The first – great – presentation was presented by Karol Draszawka, who with extraordinary diligence and enthusiasm discussed the intricacies of AlphaGo algorithm. Thanks to Karol, we learnt about the details and structure of five neural networks created by Google to defeat the masters of Go. The presentation was based on the article “Mastering the game of Go with deep neural networks and tree search“.

The second speaker was Piotr Orzechowski from Trineo company from Gdynia. He gave us an overview of the tasks for which image recognition algorithms are used in the security domain.

Our next meeting will be held on 24th of April 2017. See you!

16th meeting – abductive reasoning and Answer Set Programming

During our 16th meeting on 20th March, Robert Różański presented his PhD dissertation summary.

Subject of his dissertation is „Automation of metabolism research using abstraction logic programming” which he wrote while studying at the University of Manchaster.

Robert described assumptions and results of abductive reasoning in logic programming and paradigms of Answer Set Programming in search problems.

Thanks to Robert we were also able to grasp the basics of logic programming using such tools as gringo and clasp.

Our next meeting will be held on 3rd April. The main topic for our next meeting will be Google Algorithm for Alpha Go. Feel free to join us!

15th meeting – Dataiku Data Science Studio

15th meeting of MLGdańsk group was held on 3rd March at The Gdansk University of Technology.

This time, Marcin Zadroga presented Dataiku Data Science Studio, which is a simple and user friendly tool for data analysis. Dataiku DSS out-of-the-box offers basic and more sophisticated algorithms such as Random Forrest, Logistic Regression, Decision Trees and other.

Marcin showed us how to perform common tasks like data cleaning, training of prognostic model and its validation and evaluation.

Next meeting is on 20th March, you are welcome to join us!

14th meeting – transfer learning

On 20th February we met at The Gdańsk University of Technology for 14th meeting of MLGdańsk group.

Adam Brzeski carried out the second part of the Transfer Learning workshop. We learnt how to retrain neural network model according to transfer learning paradigm. Using Inception model we were able to train it and use it for the image classification tasks.

Our next meeting is on 6th March. All ML Enthusiasts are welcome to join us!

13th meeting – MBA – Innovation and Data Analysis / transfer learning

The 13th meeting of our group took place on 6th February.

Adam Karwan talked about his experience with MBA program he participates in – MBA Innovation and Data Analysis. Adam described overall program of studies and the most interesting modules.

Second part of the meeting was dedicated to the transfer learning workshop. This part was led by Adam Brzeski. He presented the paridigm of the transfer learning and showed us how to use TensorFlow for this purpose. He outlined thoroughly the idea of image recognition with Convolutional Neural Networks.

Our next meeting and the second part of the workshop will be held on 20nd February. Feel free to join us!

12th meeting – how to simulate life

MLGdańsk group met on 23rd January.

Marcin Kuropatwiński presented his research on creating life simulation using machine learning approach. Marcin talked about his motivation and key methodologies. He presented theoretical mathematical basis for building simulation and models of single atoms. His idea is to build simulation based on micro and macroscopic data.

For our next meeting on 6th February we would like to invite all enthusiasts of Neural Networks.

7th meeting – data analysis in aviation

7th meeting of our group took place on 17th October 2016.

Adam Karwan presented summary of the Data Science in Aviation conference which he attended.

Rest of the meeting was filled with lively discussion about the directions, topics and goals of our group.

If you are also interested in data science and machine learning, feel invited for our upcoming meetings.

6th meeting – manifold learning

6th meeting of our group took place on 3rd of October 2016.

We had opportunity to take part in lecture about manifold learning, which was delivered by Krzysztof Czarnowski. He described main principles and ideas of the manifold learning algorithms.

Feel free to join our next meeting in two weeks!

5th meeting – text representations for classification tasks

5th meeting of MLGdańsk group was held on 19th of September 2016.

At the meeting we had opportunity to listen to Karol Draszawka who delivered talk about modern methods of texts’ representation for categorization tasks. Karol thoroughly described ideas and methods e.g. BOW, n-grams, CBOW, convolutional neural nets, additionally he presented current trends in development and usage of those methods.

Next meeting is in just two weeks on 3rd of October, feel invited!

MLGdańsk

Welcome on the website of Machine Learning Gdańsk group!

We are a group of enthusiasts of data analysis, machine learning, natural language processing and artificial intelligence.

We meet biweekly, on Mondays at 6 pm. at the Gdańsk University of Technology. During our meetings, we share knowledge and talk about interesting issues.

If you are interested in joining, please contact us!

phone: 502 625 281 (Szymon), 608 821 759 (Marcin), 502 721 475 (Adam)