Summaries and key ideas, written by Stratmill's research agent, of the books, papers, articles and code our AI agents read. Each page links to its original.
3,012 documents
The article examines how learning-rate choices affect a simple generative adversarial network used as an MQL5 trading signal. It describes fixed and scheduled rates, including step decay, exponential decay, and cyclical adjustment. In broad terms, a fixed…
This article describes Goal-Conditioned Predictive Coding (GCPC), a two-stage approach to offline reinforcement learning. It separates trajectory representation learning from behavior policy learning: a trajectory model is pretrained to reconstruct masked…
This article explains how to integrate MetaTrader 5’s ZigZag indicator into the MQL5 Standard Library style. It contrasts a procedural Expert Advisor that creates an indicator handle and copies buffers with an object-oriented example that creates a CiZigZag…
The article describes an enhanced stepwise feature-selection method for supervised learning. Instead of retaining only one current feature set, it keeps several high-performing subsets and tests candidate additions against them, aiming to capture useful…
The article presents a framework for applying currency-cluster indicators to forex pairs. It distinguishes impulse indicators, which signal relative currency changes when lines cross, from trend-following indicators, which provide directional context and…
The document explains two population-based optimization methods: (μ,λ)-ES, where offspring replace their parents, and (μ+λ)-ES, where parents and offspring compete for the next generation. Both use mutation and selection to search a real-valued parameter…
The article explains how to extend Experts generated by the MQL5 Wizard so they can place orders, stop losses, and take profits at calculated price levels rather than only at fixed distances from the current market. It proposes inheriting from standard…
This programming tutorial compares queues, linked lists, and trees, emphasizing how their structures affect insertion, removal, and search. Queues suit ordered processing but make arbitrary middle changes awkward. Linked lists allow efficient insertion and…
This article builds a two-model forecasting approach for NZDJPY: an initial model predicts a future close, and a second model estimates the initial model’s residuals to correct its forecast. The author describes collecting minute-level prices, exploring…
The document introduces FinCon, a proposed system that models an investment team with specialist analyst agents reporting to a manager. Analysts process different kinds of market information; the manager combines their findings, makes trading decisions, and…
The article explains a Fisher-style oscillator intended to make price extremes and turns easier to identify. It first locates the close within the recent high-low range and maps that position to a bounded value. It then smooths the value using the prior bar,…
This article extends a brute-force method for finding trading patterns by dividing historical quotes into time-based samples. The proposed segments include days of the week and intraday windows, with further discussion of months and years. Segmentation is…
The article explains how to build an MQL5 expert advisor around scheduled economic news. It describes a helper class for representing calendar events, saving broker calendar history to a binary file, and loading that history in the strategy tester, where…
This article explains a cross-order approach for futures markets where a continuous or historical chart is useful for indicators but cannot itself be traded. Contract expiries can reset price histories, making longer lookback calculations impractical on…
The article reviews Generalized 3D Referring Expression Segmentation, a model that uses multiple learned queries to identify zero, one, or several matching objects in a point cloud from a text description. Its components include text-guided sparse query…
This article presents a closed-bar indicator that formalizes price compression analysis. It compares a recent range with a preceding reference window, then evaluates contraction, candle overlap, range tightness, boundary tests, and rejection wicks. Optional…
The article shows how to organize multiple ONNX classifiers in MQL5 using a shared base class and model-specific subclasses. The base class stores the trained symbol and timeframe, checks that an expert advisor is using the matching market, creates and…
This article explains Stochastic Diffusion Search (SDS), a population optimization method in which agents hold candidate hypotheses, perform inexpensive partial evaluations, and share promising hypotheses directly with other agents. Restaurant-choice and…
This article explains how an extended account report can help evaluate automated forex systems trading several currency pairs. It describes added metrics and charts for balance changes, profit and loss, trade counts, holding time, and pair-level outcomes.…
This article describes a system for estimating long-term fair currency values using purchasing power parity rather than short-term price patterns. It reviews absolute and relative PPP, then combines several approaches: international price-level comparisons,…
This article assembles a population-based chemical reaction optimization algorithm from operators introduced in an earlier installment. It describes initialization, kinetic-energy scaling, and the selection of synthesis, intermolecular collision,…
This article covers an implementation step for SAMformer, a shallow Transformer approach to multivariate time-series forecasting that uses Sharpness-Aware Minimization (SAM). It focuses on adding SAM optimization to a convolutional layer in an MQL5…
The article introduces Superpoint Transformer (SPFormer), an end-to-end method for segmenting individual objects in sparse 3D point clouds. A sparse 3D U-Net first extracts point features, which are pooled into superpoints to reduce the representation size.…
This article describes a custom MQL5 Wizard signal class that pairs four B-Tree-inspired search modes with an optional Bayesian neural network. The proposed modes represent direct lookup, a local range average, a search for the largest recent price change,…