Research areas at the Core
Ekkono is a knowledge-driven company built on cutting-edge research on edge machine learning. The uniqueness, and the result of seven years of machine learning research at the University of Borås in Sweden, is the ability to do incremental learning on streaming sensor data onboard the device. Being on the forefront of research, using it in our daily corporate life, is crucial to stay on top our product development. At Ekkono we have leading Data Scientists in the fields of Incremental Learning and Energy Efficient Algorithms. Our research areas stretch from Sensor Data, to Machine Learning Algorithms and Resource-Constrained Devices.
Federated Learning
Federated Learning is a machine learning technique to train a global model from different clients or devices while keeping the data decentralized. A central server is in charge of coordinating the transfer of the models’ data between the server and the clients, to update the global model while keeping the data private without leaving the device.
Conformal Prediction
The conformal prediction framework offers an alternative method for constructing and evaluating predictive models that appears better suited than traditional predictive methods in applications where thorough verification is crucial. Whereas traditional predictive models output so-called point predictions—a single-valued best-guess prediction for the value of the dependent variable—conformal predictors output multi-valued prediction regions that represent a range of likely value assignments for the dependent variable, constrained by its domain.
Energy Efficient Algorithms
While more and more devices are being connected, they consume more and more energy. What happens when you also run machine learning (ML) on these devices? How can we design ML models that are as environmentally friendly as possible from the start? And better – how can we design algorithms that also train devices to be better and smarter and therefore minimize their CO2 footprint? At Ekkono we have a sustainable mindset from start to finish. We focus on making our ML algorithms as energy efficient as possible, to make sure our set of algorithms and software library does not consume more energy than necessary.
Incremental Learning
Incremental learning consists of training a machine learning model from streaming data to continuously learn over time as more data and insights are being gathered. The model is being built and updated as the data arrives, and predictions are made at any point in time. Traditional machine learning algorithms in an IoT setting would collect data from many different devices to build a generic algorithm that is representative of all devices.