AI @ Industry

Hackathon

powered by

Siemens AI Lab


  Join us for the final pitches, award ceremony & networking!  


   Register now for free  




24 - 27 April 2019


Join forces with one of the global industry leaders and tackle AI challenges that shape our future.


Research in Artificial Intelligence is advancing rapidly and already having a visible impact on our digital lives. But how can we make a difference in the physical world? How will we implement the full potential of modern AI technologies in industrial applications? And will you be part of the journey?

The Siemens AI Lab Munich is partnering up with the MindSphere Application Centers around the globe to bring their most interesting industrial AI applications to you for a 72-hour Hackathon. Apply to your favorite project track and solve relevant challenges in machine learning, deep learning, natural language processing, knowledge graphs, semantic modeling, computer vision, signal processing, operations research and other fields.

This is your opportunity to join forces with Siemens domain experts and data scientists, gain valuable insight into exclusive data and contribute to real AI-based solutions by bringing your expertise to the table. And also, our MindSphere Application Centers are currently looking for talents to join their teams in exciting locations like Sevilla, Dubai or Singapore.

So, pack your brain, your laptop and your team spirit, come to our AI Lab, and start hacking!

When: 24 – 27 April 2019
Where: Siemens AI Lab @ Mindspace Viktualienmarkt


APPLICATION CLOSED

What is it all about?

The Siemens MindSphere Application Centers have been designed to develop digital solutions and services as well as industrial applications. With their expert teams of software developers, data specialists and engineers, they co-create digital offerings together with Siemens customers in 50 locations across the globe. Therefore, they are at the forefront of bringing modern data-driven AI technologies into the industrial applications of tomorrow.

For this Hackathon, the MindSphere Application Centers have joined forces with the experts from the Siemens AI Lab to prepare their most exciting AI-based challenges as 72-hour hacking tracks (Wednesday evening to Saturday evening). When you register for the event, you can apply directly to your favorite track that matches your expertise and interest.

This event is for all AI and data enthusiasts with experience in machine learning, deep learning, natural language processing, knowledge graphs, semantic modeling, computer vision, signal processing, operations research or other related fields. So, people like you who want to improve water quality, optimize traffic systems, implement speech recognition, analyze knowledge graphs, increase energy efficiency, try something new, and have an awesome time!

Together with your team captains, you will be working on their provided challenge in a team of 4-8 like-minded people starting on Wednesday evening 5pm. After three days of coding, analyzing and innovating, on Saturday evening, the final event will take place at the Siemens HQ at Wittelsbacherplatz, where your team will present their results in a final pitch in front of an exquisite jury as well as a wider audience. And in the end, you can win awesome prizes and maybe even get a job opportunity.

We’ll take care of you! - food and drinks all for free.

Wanna know more: FAQs

Email us in case you have any question: team@ai-hackathon.com


<Clck for next picture!>
Click image for next picture.


 

Great Challenges are waiting for you - Make your choice!
Actually it's an easy choice: All challenges will be a unique experience! Enjoy!

Track 1: Object Detection in Traffic Video Data with Edge Devices

Hosted by: Urban Infrastructure MAC, Singapore

#What it is about:

The traffic conditions in many developing countries are often characterized by its chaotic nature, where road lanes are generally interpreted as merely a suggestion. With high densities and limited budgets, there is an increasing demand for low-cost intelligent solutions that can reduce traffic congestion. In this track, the objective is to design a lightweight traffic condition sensing solution using artificial intelligence methodologies that can run in real time on low-cost, scalable hardware (e.g. edge devices, low-end accelerators).

#What you need (at least 1):
  • Understanding in Neural Networks and Deep Learning
  • Hands-on experience in developing deep neural networks using popular frameworks, preferably in Tensorflow, PyTorch, Keras, MXNet
  • Or alternatively hands-on experience in developing applications in OpenCV
  • Experience or interest in architectures such as (Mask) R-CNN, SSD, YOLO
  • Experience or interest in computer vision techniques, especially object detection
  • Experience with low level programming (e.g. C/C++), experience with AI/ML hardware accelerators, multicore/parallel programming, real-time systems
  • Knowledge of methods and techniques to optimize (e.g. quantize, sparsify) neural networks
#What you can get:

Promising candidates can look forward to potential full-time and internship opportunities at the Siemens Digitalization Hub in Singapore.

Track 2: Road Traffic Forecasting from Sensor Data

Hosted by: Intelligent Traffic Systems Digitial Labs, Munich

#What it is about:

Many cities suffer from road congestion due to ever increasing annual traffic volumes. In this project you will leverage the power of artificial intelligence to develop algorithms that can predict road traffic over different forecast horizons. Large amounts of traffic data will be analyzed using state-of-the art time-series models such as deep learning.

#What you need (at least 1):
  • Experience in time-series analysis, modeling and forecasting, statistics,
  • Experience with machine learning and deep learning (particularly RNNs)
  • Experience with python and machine learning frameworks
  • Knowledge in application design, frontend-design (dash /bokeh), AWS is a plus
  • Traffic-domain knowledge is a plus
#What you can get:

Promising candidates can look forward to potential full-time and internship opportunities at the ITS Digital Labs in Munich.

Track 3: Process Optimization in Production Facilities

Hosted by: Energy Efficiency MAC, Sevilla

#What it is about:

Energy consumption is a main cost driver in most industry sectors, and it correlates directly with a factory’s carbon dioxide emission. The potential for optimizing processes subject to minimizing energy consumption is therefore extremely high, but rarely taken into account in process planning and design. Our goal in this track is to develop and implement a tool which does this and at the same time guarantees that required outputs and deadlines are met. We will directly see the impact of your work since we will test our tool on actual operations data.

#What you need (at least 1):
  • Practical experience in Operations Research, especially modelling and solving discrete optimization problems (MILPs, heuristics etc. ) or nonlinear problems
  • Programming skills (e.g. C, C++, C#, Python)
  • Knowledge about optimization; production planning or similar problems is a plus
  • Experience with large-scale data lifecycle management, distributed data management systems and big data analytics is a plus
  • Experience with software architecture and visualization
#What you can get:

Promising candidates can look forward to potential full-time and internship opportunities at the Energy Management MACs in Munich and Sevilla.

Track 4: Energy Efficiency in Air Compressors

Hosted by: Energy Efficiency MAC, Sevilla

#What it is about:

Compressed air production is a main cost driver in many industries and can amount up to 50% of the total production cost. In addition, any improvement of its efficiency directly impacts a facility’s CO2 emissions. The potential for optimizing the efficiency of compressed air production is therefore extremely high, but rarely taken into account due to the main focus on production yield. Our goal in this track is develop a solution which achieves this and at the same time guarantees that production targets and deadlines are met. Since we build the solution based on real field data we will directly see the impact of our solution.

#What you need (at least 1):
  • Experience in Machine Learning
  • Experience with AI frameworks (e.g. TensorFlow, Watson, Einstein, H2O, …)
  • Programming skills (e.g. Python, R, Scala)
  • Experience with large-scale data lifecycle management, distributed data management systems, big data analytics, data quality and anomaly detection in time series.
  • Profiles not so close to programming but with knowledge about compressors, electric drives, hydraulics or air flow behavior are also welcome.
#What you can get:

Promising candidates can look forward to potential full-time and internship opportunities at the Energy Management MACs in Munich and Sevilla.

Track 5: Semantics Extraction from Service Reports

Hosted by: Industry Services MAC, Erlangen

#What it is about:

If you want to generate knowledge graphs from natural language, this is your chance. Currently, service technicians report their work in service reports, but the gathered knowledge isn’t used in any way afterwards. We propose the challenge to process these reports and make the knowledge searchable with keywords. Globally generated service reports may lead to a recommender system that is able to propose solutions, required competences and required tools for a service.

#What you need (at least 1):
  • Experience with knowledge representation, semantic web / linked data or knowledge graphs
  • Experience with relevant technologies, e.g. RDF(S), OWL, Semantic Mapping (R2RML), Sparql, Property Graphs, Cypher, Gremlin, …
  • Experience with Natural Language Processing is a plus
  • Experience with Machine Learning is a plus
#What you can get:

Promising candidates can look forward to potential full-time and internship opportunities at the Industry Services MAC in Erlangen.

Track 6: Speech Recognition for Air Traffic Control

Hosted by: Airports MAC, Dubai

#What it is about:

Speech recognition in air traffic communication is a very challenging task due to noisy environment conditions and high speech rate of voice. Our solution will analyze unstructured raw audio of ATC communication and provide a service of real time text transcription of audio communication of ATC. Deep learning, signal processing and NLP will be used to model the relationships between voice and text and extract relevant information such as speaker id (controller, pilot) aircraft arrival information (call-sign, clearance) and security alerts.

#What you need (at least 1):
  • Understanding in Neural Networks and Deep Learning
  • Hands-on experience in developing deep neural networks using popular frameworks, preferably in Tensorflow, PyTorch, Keras
  • Experience or interest in speech recognition applications using relevant architectures like RNNs
  • Experience or interest in Natural Language Processing
  • Experience or interest in front-end development (Angular, React) would be a plus
  • Experience with AWS is a plus
#What you can get:

Promising candidates can look forward to potential internship opportunities at the Airport MindSphere Application Center in Dubai.

Track 7: Water Quality Estimation from Hydraulic Network Data

Hosted by: Water & Waste Water MAC, Manchester

#What it is about:

'Clear' thinking - no longer going with the flow
With aging pipes and ever tighter regulation, maintaining supply of clean, clear water to consumers is a tough challenge for water companies.

Discoloured water is caused by the build-up and movement of material in a water network. Turbidity sensors estimate the amount of material in the water and are being increasingly installed by water companies at key locations in their networks. Whilst discoloured water is generally recognised to be the result of hydraulic variability in the water network, extensive investigation of these processes in a live distribution system at the city scale has been limited by a lack of data.

Here we provide a pioneering dataset including network schematics, flow, pressure, and turbidity from more than 50 sites across a UK city. Can you use the hydraulic dataset (flow, pressure) and network data (pipe material) to predict the turbidity response of the system and ensure clear water for the residents?

We are an expanding team based in Manchester UK, always on the lookout for talented data scientists and other developers with, amongst others, these types of skills.
  • Familiarity with electrochemical and optical sensors
  • Machine learning cloud platforms (AWS, Azure)
  • Time series forecasting/prediction
  • Feature engineering
  • Visualisation
#What you need (at least 1):
  • Experience in time-series analysis, modeling and forecasting, statistics
  • Experience with machine learning and deep learning (particularly RNNs)
  • Experience with python and machine learning frameworks
#What you can get:

Promising candidates can look forward to potential full-time and internship opportunities at the Water & Waste Water MAC in Manchester.

Track 8: Water Pipes Burst Detection & Localization

Hosted by: MAC Abu Dhabi, UAE

#What it is about:

Pipes are used worldwide to transport water from main reservoirs to several entities. However, monitoring the integrity of these huge networks in terms of identifying bursts and localization them is a major concern for water utilities operators. Not only do bursts cause water loss, but also energy consumed in pressurizing water as well as harming customers’ satisfaction. When a burst takes place within the pipeline network, a pressure transient wave(s) is/are propagated through the network, these transients are considered strong burst indicators. Furthermore, they are used for localization. This project aims at using pressure readings to detect and localize bursts along pipelines.

#What you need (at least 1):
  • Signal processing techniques
  • Pattern recognition/ template matching
  • Machine Learning on small datasets
  • Transient wave detection
  • Correlation analysis
#What you can get:

Promising candidates can look forward to potential full-time and internship opportunities at the UAE MAC in Abu Dhabi.





Please read our FAQ
to learn a lot more about the hackathon.


A great hackathon experience is waiting for you!

APPLY NOW!



Imprint | Privacy Notice | Terms & Conditions | Contact