MachineStreamLearning
for Quants & Traders
Real-time, end-to-endopen box AI
What is Stream Learning?
a solution to the end-to-end data lineage problem
Traditional Machine Learning
Trying to create dynamic outputs from static inputs is logically inconsistent
Disc Stream Learning
In-flight AI ensures zero information loss or latency; links cause to effect
What is Disc?
a solution to the end-to-end data lineage problem
Explore Data Streams
Select from a catalogue of pre-curated realtime data feeds.
Integrated analysis generates cursory metrics and a view of a stream’s behavioural characteristics.
Mathematical Modelling
Derive any number of real-time variables to define a mathematical model for each streamed input.
Integrated backtesting and live model testing to improve the predictive performance of models.
Build Trading Bot
Use Disc’s open-box AI architecture to build a decision-making bot rule by rule.
End-to-end data lineage allows you to link cause and effect - directly to an individual variable.
How Does Disc Work?
no need to do the engineering - just do the maths
below is an example
Stochastic Model
Suppose a simple Gaussian distribution to model the price of a given stock which deviates around a mean value
Real World Observations
Observations are made in real-time; we can therefore evaluate the probability of having made those observations
Trading Strategy Execution
The vectorised probability for the next observation can be formed; this is used to build an N-dimensional decision model
Want to try Disc?
we’re currently in alpha testing
reach out if you’re a Quant or Trader