The Solution for Industries

We present several important projects from our technologies developed by our R&D team. For most of our projects, the core AI engines have been developed and ready to be productised using open resources by top Silicon Valley labs. For more details or other projects, please contact us.

Some works

We create the best Solution
to solve some hard problem


Our STADLE platform is a paradigm-shifting technology combining Machine Learning (ML) and decentralized system capability to provide scalable, versatile, and secure AI services. STADLE stands for Scalable, Traceable, Asynchronous, Decentralized federated LEarning.

  • Federated Learning (FL) solves the problems of privacy and communication load, which commonly appear in ML systems. FL does not require users to upload raw data to cloud servers.
  • Privacy: FL improves the privacy-preserving aspect of AI systems by not collecting data in the cloud while producing collective intelligence based on uploaded user ML models.
  • Communication load: The amount of traffic generated by FL dramatically decreases from classical AI systems due to the difference in data type exchanged.

Our STADLE platform enhances the capability of FL by incorporating decentralized architecture.

  • Scalability: Decentralized FL servers in STADLE realise the load-balancing to accommodate more users.
  • 5G-friendliness: The delay in communication to obtain collective intelligence can be dramatically reduced by employing decentralized FL servers located at edge servers.
  • Traceability: Our platform has the performance tracking capability that monitors and manages the transition of collective intelligence models in the decentralized system.

Privacy-preserving Ad Delivery

The targeted advertisement is facing challenges related to data privacy regulations.

TieSet’s Federated Learning platform (STADLE) and Transfer Learning technologies realize effective personalized ad delivery while preserving the privacy of users’ browsing data.


Training robots for new tasks consumes a lot of resources and time.

Fast improvement in robots’ performance on diverse tasks will be observed when they are trained in TieSet’s Federated Learning platform (STADLE) and Leaning Synthesis technologies that create collective intelligence.


No one wants to share their healthcare data, but everyone wants to know their health diagnosis in detail.

TieSet provides AI healthcare systems that analyze behavior patterns without transfers of personal health data by its Federated Learning platform (STADLE).


Automatic interpretation of voice mails, long emails, and video footage saves a lot of time to understand the content so we can focus on “actual” tasks.

Personalized Federated Learning platform (STADLE) designed by TieSet is a solution to receive benefits of an intelligent virtual secretary giving you the interpretation while not giving up your data privacy.

Smart City

Automatic control of room environments enhances our well-being at workplaces and homes.

TieSet’s smart city system controls the environment using the shared intelligence learned through its Federated Learning platform (STADLE) and Transfer Learning technologies.

Smart Energy and Smart Buildings

The Internet-of-Things (IoT) paradigm can be scaled up to the level of civil, commercial, and residential infrastructure. The prevalence of smart-grids, smart-home technologies, and Building Automation Systems make energy optimization within the reach of distributed AI approaches.

The Smart Grid is a networked system of sensors, controllers, and machines that share data to collectively optimize for a global objective. Smart Grids can intelligently manage energy usage to save costs, improve safety, and ensure human comfort. However, the availability of large amounts of data presents inherent privacy risks and computational costs.

Federated Learning can be used to keep data private while sharing intelligence to collectively increase performance. Additionally, Deep Reinforcement Learning can be used to provide adaptive controllers that tailor their operations to changing environments and individual requirements of machines in the smart grid. The STADLE platform combines these two approaches to provide private, adaptive, and distributed AI.

According to Statista[1], Smart City technologies will account for $679.5 billion of spending by public and private organizations. Whereas the largest current market for Smart Buildings is North America, the fastest-growing region is the Asia Pacific with 23% year-on-year growth [2]. Such an established market with a dynamic growth profile presents an attractive target for quantum leaps in technology provided by the STADLE platform.



Manufacturing Robotics

It is a framework for low risk of injury, data privacy, and smart adaptive collaborative devices. Smart devices can be trained and learn new tasks while keeping the patient’s data private and without posing a threat when interacting with humans. The framework is generalized to support all types of devices.


It is a framework for low risk of injury, data privacy, and smart adaptive collaborative devices. Smart devices can be trained and learn new tasks while keeping the patient’s data private and without posing a threat when interacting with humans. The framework is generalized to support all types of devices.

Semantic Extraction

Emotion detection is a crucial factor in human-computer interactions, including dialog systems. Speech emotion recognition can be described as predicting the emotional content of speech and classifying speech according to one of several labels (i.e., happy, sad, neutral, and angry). Various learning methods have been applied to increase the performance of emotion classifiers. Since data is sourced at low cost via local devices and users’ privacy is essential, federated learning is a suitable choice as network architecture. In this project, we propose an advanced federated learning framework for emotion detection applications. In addition, since the voice pattern and emotional state of each user are different, the locally generated data will be highly non-IID. To address this problem, we perform personalized federated learning to maintain a specific local model for each user.

Transitioning from big data to collective intelligence. Moving towards the Internet of Intelligence.