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.
5,018 documents
This article explains Bacterial Foraging Optimization (BFO), a population-based method for numerical optimization inspired by bacterial chemotaxis. Candidate solutions move through a search space using swimming and tumbling, while swarming encourages…
The Exchange Market Algorithm (EMA) is presented as a population-based method for numerical optimization, inspired by traders with different levels of success. It ranks candidate solutions into elite, middle, and beginner groups. The elite preserve their…
The article describes a workflow for taking a calibrated machine learning model from a Python research pipeline into MetaTrader 5 for execution-aware backtesting. Python exports the ONNX model, calibrator parameters, feature specification, and combinatorial…
The article explores adding Python-trained deep learning models to three existing Expert Advisors based on causal network analysis, stochastic model optimal control, and Nash game theory. It describes using historical EURUSD prices to train a convolutional…
The article presents a computer-vision workflow for classifying market chart screenshots with a convolutional neural network. It defines buy and sell image categories around directional price moves and daily extremes, then describes collecting hourly…
This article proposes a custom MQL5 money-management class for traders whose decisions depend on recurring price patterns rather than volume accumulation. It converts price changes into a discrete sequence of up, down, or flat states. A suffix automaton…
The article presents a workflow for using a Latent Gaussian Mixture Model to identify hidden structure in technical-indicator data. It introduces mixture components and latent assignments, then describes fitting the model with expectation-maximization, which…
The article introduces FinAgent, a proposed market analysis agent that combines text and visual information with historical memory, reflection, and trading tools. Its five-part design covers market analysis, retrieval from memory, short- and long-horizon…
This article turns a prior decision-forest classification experiment into a modular MetaTrader Expert Advisor. The classifier assigns bearish, neutral, or bullish labels from completed-bar features; the author uses “regime” to mean a persistent directional…
This article explains how to turn a point cloud and its pairwise distances into a Vietoris–Rips filtration for topological data analysis. Vertices enter at zero, edges enter at their endpoint distance, and triangles enter at the longest of their three edges.…
This article adapts category theory functors to generate short-term S&P 500 trading signals from linked economic calendar data. It models data points such as retail sales, 10-year auction yields, PMI, and CPI as a graph, then uses neural networks to map…
The article introduces Deep Deterministic Policy Gradient (DDPG) as a reinforcement-learning filter for signals derived from moving-average and stochastic-oscillator patterns. It contrasts DDPG’s continuous-valued action output with discrete buy, sell, or…
The article describes an MQL5 trading model that represents three qubits as a joint eight-state system. It outlines complex state amplitudes, probability measurement, rotation gates driven by market features, controlled phase gates based on feature…
The article introduces Gaussian process regression as a Bayesian, kernel-based approach that predicts a distribution over possible function values rather than only a point estimate. It explains how a prior encodes assumptions about function shape, how the…
The article introduces recurrent neural networks as a way to process price sequences while carrying information forward between time steps. It explains why a fixed input window can miss events outside the selected span, then presents an LSTM block with…
The article explains natural transformations through two functors that map normalized changes in ATR and Bollinger Bands to price-range changes. Indicator changes are discretized into 21 bins from −100% to 100%, while price-range changes are stored as…
This article introduces three deterministic optimization methods available through ALGLIB in MetaTrader 5: BLEIC, L-BFGS, and NS. It explains objective functions, adjustable parameters, and box, equality, and inequality constraints, then focuses on practical…
This article’s first installment explains how to bring parts of the Kronos candlestick foundation model into native MQL5. It covers offline export of PyTorch parameters to binary float files, candle preprocessing, and the tokenizer encoder, which compresses…
The article describes custom Python implementations of DeMarker and price Envelopes, then turns combinations of their signals into features for a recurrent neural network using a white-noise kernel. DeMarker is calculated from smoothed increases in highs and…
The article explains linear discriminant analysis (LDA) as a way to project labeled data into a lower-dimensional space that emphasizes class separation. It contrasts LDA with PCA, which prioritizes overall variance, and QDA, which estimates separate class…
This article explains why classifier probabilities need calibration before they are used for trade sizing. A model can rank opportunities well while overstating their true win rates; that error can flow into probability-based sizing and Kelly-style…
The article introduces gradient boosted decision trees and explains how XGBoost and LightGBM build an ensemble sequentially, fitting trees to reduce the errors of earlier predictions. It covers practical distinctions from methods such as logistic regression,…
This article describes an Expert Advisor that combines an ONNX price classifier with SMA and EMA trend filters. Predicted direction is used to trigger trades only when price also lies on the corresponding side of both moving averages. The system periodically…
The article explains Ebola Optimization Search Algorithm (EOSA), a population-based metaheuristic inspired by disease transmission. It maps close contact to local exploitation around a strong solution and travel to exploration across the search space, while…