Parallel Parameter Runs for AutoML Workflows in BigQuant

Summary

The document outlines a way to run multiple parameterized tasks in BigQuant, a low-code platform where a visual workflow is represented as a graph of modules. A parallel mapping function can dispatch different inputs to a function, and the same pattern can pass distinct parameters into graph modules so several calculations run concurrently. BigQuant’s custom-run module is identified as the mechanism for connecting the parameter sets to the workflow.

This describes parallel experiment execution rather than an AutoML or AutoDL method: it does not specify automated model selection, tuning criteria, data handling, or evaluation procedures. A small illustrative example shows a function applied to two inputs, but no performance measurements or trading results are reported. The note says reinforcement learning is still under research and is not expected to be available soon, which limits the scope of the platform capabilities described.

Key ideas

• BigQuant represents visual low-code workflows as graphs of modules.
• Parallel mapping can run a function across different inputs at the same time.
• The custom-run module can pass different parameter sets into graph modules for concurrent calculations.
• The document explains parallel execution, not an AutoML model-selection or tuning process.
• Reinforcement learning is described as still under research and not yet available.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.