Modern ML for Trading
The full gradient-boosting-to-neural-operator toolkit - ensembles, AutoML, probabilistic forecasting, distillation and pruning for latency, plus physics-informed architectures like Neural ODEs and Fourier operators.
用于收益方向预测的 XGBoost：类别不平衡与决策阈值
收益方向分类器属于不平衡问题，0.5 不是正确的决策阈值。本文在加密货币数据上比较 scale_pos_weight、焦点损失和精度约束阈值优化，并介绍 XGBoost、LightGBM 与 CatBoost 之间的工程差异。
集成方法：结合弱学习者以获得稳健的 Alpha
本博客未涵盖有关交易集成的三件事：为什么协方差项（而不是模型计数）在集成误差中占主导地位，堆栈如何作为具有折叠元特征的信号组合层工作，以及跨多个信号的周转净额是否真正降低了执行成本。
用于系统化交易流程的 AutoML
自动特征生成（tsfresh、Featuretools）、预算感知的模型搜索（FLAML 的节约成本优化器）以及 WorldQuant 公式化 alpha 工厂 - - 这篇博客的搜索与过拟合系列尚未覆盖的研究流程部分，以及该系列自身结果对它们的启示。
Multi-Task Learning for Simultaneous Price, Volume, and Volatility Prediction
Does jointly predicting return, volume, and volatility actually help? Measuring loss-balancing schemes and diagnosing negative transfer through gradient cosine similarity - with a classical baseline and purged walk-forward folds.
交易机器人保护中的异常检测：从Z-Score到Transformer
哪些异常检测方法在加密货币算法交易中真正有效，如何构建级联防护架构，以及为何这是算法交易不可或缺的基础。
Scoring Probabilistic Forecasts: CRPS, PIT Calibration, and DeepAR
How to evaluate a predictive distribution honestly - CRPS as a proper scoring rule, the PIT histogram as a calibration diagnostic, and DeepAR sampling in GluonTS.
Irregular Time in Tick Models: Continuous-Time Encodings vs. Plain Positional Embeddings
Sequence models fed tick data still assume regular spacing. Three ways to tell a Transformer when a tick actually happened - learnable-timescale continuous encoding, ODE-RNN latent state, and delta_t as a plain feature - and the ablation that decides between them.
一整天的内容能胜过十分钟的内容吗？ Flash 注意力和序列长度问题
Flash Attention 使 23,400 步的交易上下文无需计算 - - 平铺、在线 softmax 和 N^2 d^2 / M 的 IO 界限。更长的上下文是否使模型更好是一个单独的问题，它必须被测量，而不是断言。
Knowledge Distillation: Compressing Trading Models for Low-Latency Deployment
The blog's standing answer to the accuracy-vs-latency tension is a two-stage fast/slow split. Distillation is a different answer: train one small model to mimic the ensemble. The KD loss, temperature, born-again nets, early exits for a variable latency budget, and the distill-to-FPGA pipeline - plus the measurements that would decide whether it beats the two-stage split.
Model Pruning for Low-Latency Trading Inference
Magnitude and structured pruning, the Lottery Ticket Hypothesis, movement pruning, distillation and 2:4 sparsity - the methods behind shrinking a trading model, and what still has to be measured before any of it ships.
非参数价格建模的高斯过程
金融时间序列的核设计 - - 母体粗糙度、局部周期成分、光谱混合 - - 以及作为不需要验证集的正则化器的边际似然。加上在任何东西可以交易之前仍然需要衡量的诚实清单。
认识与对立：衡量收益模型不知道的东西
本博客的每个 position sizing 规则都把不确定性坍缩成单一纯量。MC Dropout 和深度集成将其拆分为模型无知与市场噪声 - - 这两者值得不同的 position size。
Neural ODEs: Does Continuous Time Beat a Delta-Time Feature?
Neural ODEs, Neural SDEs and continuous normalizing flows for irregularly-sampled market data - and the one ablation that decides whether continuous dynamics are worth their solver cost.
Hamiltonian Neural Networks: Does a Financial System Conserve Anything?
Hamiltonian Neural Networks are provably stable - but stability is worthless if the conserved quantity does not exist. Learning a scalar H via autograd, symplectic integration, and the falsification test that decides whether a financial (q, p) pair is canonical at all.
基于偏微分方程的金融建模的傅立叶神经算子
算子学习映射整个函数空间，而不是点 - - 傅里叶神经算子如何在频率空间中参数化 PDE 解算子，为期权定价购买什么，以及它的哪些主张仍然需要在真实硬件上测量。
