Fides et Ratio

Welcome to Scientific Machine Learning

If you have time-series anywhere, Scientific Machine Learning measures it directly and exactly, with measurements available immediately, without model building, model training, or model bias, that inevitably lag behind when decisions must be made---in the present. The first step is the SML-Training Viewer (free) that consists of core training on Scientific Machine Learning (SML) measurements to be well-informed. On completing the training the full SML-App Viewer and Decision Machine can be downloaded. You can begin to explore your own time-series. Finally, we cordially invite you to share your questions, insights and discoveries with the Scientific Machine Learning community at Fides et Ratio.

Features

The science-of-counting is science when counting in units. This is achieved by counting up to the last full measure, which ensures that multiple measurers, given the same unit of measure, will get the exact same measurement. Any leftover by definition is less than a full unit of measure, and is ignored.

Scientific Measurements

Scientific Machine Learning (SML) integrates the science-of-counting with time-series data to define deductive, combinatorial scientific measurements, including, exact discrete probability distributions, the energy entering or exiting the system, temperature, dissipation, and so on. Scientific Machine Learning generates real-time, model-free, vertical independent measurements from time-series, also known as pure measurements.

Human Decisions

Human decisions operate in the non-equilibrium reality measured by time-series. Because human decisions can respond to external influences, control decisions are also human decisions. Mechanical decisions, on the other hand, forbid external influences and assume that the statistical future is the same as the statistical past, equivalently, no energy is allowed to enter or exit. Human decisions cannot be expressed as fixed-rule procedures on data.

Transparent, Accessible and Inclusive

Mathematically transparent deductive measurements that are available for processing in the cloud. No models and vertical agnostic. Anybody can be a data-scientist by taking scientific measurements and thinking about their meaning. Fides et Ratio will host global community discussions (github) and help to organize what is learned.

Our Blog

Lessons learned and directions to explore.

To Give & To Receive: Scientific Learning

Machine Learning

12

Aug

13:23

Over the past decade, the 19th century science-of-counting has been resurrected to provide a combinatorial derivation of conventional Machine Learning that uniquely generalizes statistics to probability theory, allows energy to enter or exit the system, and processes any time-series to return a complete set of scientific (thermodynamic) measurements as deductive reality. And now with a plausible way to explain and generalize Machine Learning for time-series, we observe that “Machine” and “Artificial Intelligence” are too prominent in this case, because in the science-of-counting computers only do what they have done from the beginning: evaluate built-in functions. The derivation of the functions is human intelligence, not artificial intelligence. In the science-of-counting “scientific machine learning” will be simplified to scientific learning.

Portfolio

Maecenas tempus nec mi et placerat. Duis at odio vitae ex gravida volutpat. Nulla finibus aliquet diam ut eleifend. Mauris consequat interdum quam. Mauris dui sapien, efficitur et cursus id, molestie fringilla turpis.