Tools for assessing clustering robustness

Forge strategies through human-AI collaboration. You bring the edge, AutoForge stress-tests it, optimizes it, and tells you if it is real. See the case study: 200+ experiments, 8 phases, 1 validated strategy.

Exploring non-gradient-based learning techniques for training neural networks, using brute force parameter search and optimization methods. Includes comparison with gradient-based learning.

Universal noise model for superconducting quantum chips achieving 5.2-19.5× accuracy improvement over traditional methods through cross-platform parameter optimization.

AI-Driven Control Strategy for Differential Drive Wheeled Mobile Robots: Neural Network-Based Parameter Optimization and Real-Time Stabilization for Multi-Waypoint Navigation.

A sophisticated PDF document analysis and question-answering application that leverages advanced AI models to provide detailed responses to user queries about PDF documents.

台股盤中當沖量化交易系統 — 追蹤主力資金動向，系統性捕捉機構操作後的剩餘價差。整合即時行情、族群連動分析、自動化做空策略、參數最佳化與歷史回測框架。

Advances in the models for studying cardiovascular physiology: Using parameter sensitivity analysis (PSA) to reduce the length of the voltage protocol

Parallel optimization engine for QuantConnect LEAN algorithmic trading strategies

This repository contains the modules implementing a Machine Learning-based solution for optimizing the execution of dislib algorithms. In particular, a stacked classification model is leveraged to predict the most suitable value of the block-size parameter for the execution of dislib algorithms.

Distributed GPU-accelerated MetaTrader 5 strategy backtester and parameter optimizer. Maps each parameter combo to a CUDA thread, shards across a Ray cluster of GPU workers. Numba CUDA + Ray + Polars + Pydantic.

Optimization of parameter values means finding the best combination of the parameters that governs the model, to enable it to perform the given task with relative accuracy

Inventory-Aware Market Making Parameter Optimization in a Simulated Exchange

Unscented Kalman Filter (UKF), to enhance modeling and understanding of neural dynamics from fMRI data using Coupled Oscillator Model.

GPU-accelerated function evaluator and optimizer for parallel parameter space exploration

This project compares the effects of Ridge (L2) and Lasso (L1) regression models on clinical data.

Official implementation of DoLQ, a multi-agent LLM framework for discovering physically plausible ODEs from observational data.

Efficient reasoning under constrained compute.

Automate strategy research: turn trading ideas into code, test parameters, and validate edge with AI-driven analysis