← All Collections
🧬 15 parts

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.

  1. 01
    XGBoost for Return Direction: Class Imbalance and Decision Thresholds
    Aug 25, 2026 #machine-learning

    XGBoost for Return Direction: Class Imbalance and Decision Thresholds

    Return-direction classifiers are imbalanced problems, and 0.5 is the wrong decision threshold. Comparing scale_pos_weight, focal loss, and precision-constrained threshold optimization on crypto data — plus the engineering differences between XGBoost, LightGBM, and CatBoost.
  2. 02
    Kaedah Ensembel: Menggabungkan Pelajar Lemah untuk Alfa Teguh
    Aug 3, 2026 #machine-learning

    Kaedah Ensembel: Menggabungkan Pelajar Lemah untuk Alfa Teguh

    Tiga perkara mengenai ensembel perdagangan yang blog ini tidak meliputi: mengapa istilah kovarians — bukan kiraan model — mendominasi ralat ensembel, cara tindanan berfungsi sebagai lapisan gabungan isyarat dengan ciri meta luar lipatan, dan sama ada jaringan pusing ganti merentas banyak isyarat sebenarnya mengurangkan kos pelaksanaan.
  3. 03
    AutoML untuk Saluran Paip Dagangan Sistematik
    Jul 30, 2026 #AutoML

    AutoML untuk Saluran Paip Dagangan Sistematik

    Automated feature generation (tsfresh, Featuretools), budget-aware model search (FLAML's cost-frugal optimizer), and the WorldQuant formulaic alpha factory — the parts of the research pipeline this blog's search-and-overfit arc never covered, and what the arc's own results say about them.
  4. 04
    Multi-Task Learning for Simultaneous Price, Volume, and Volatility Prediction
    Aug 15, 2026 #deep-learning

    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.
  5. 05
    Pengesanan Anomali untuk Perlindungan Bot Dagangan: Dari Z-Score ke Transformer
    Feb 19, 2026 #dagangan algo

    Pengesanan Anomali untuk Perlindungan Bot Dagangan: Dari Z-Score ke Transformer

    Kaedah pengesanan anomali yang benar-benar berkesan dalam dagangan algo kripto, cara membina seni bina perlindungan berlapis, dan mengapa ini adalah asas yang tanpanya dagangan algo menjadi perjudian.
  6. 06
    Scoring Probabilistic Forecasts: CRPS, PIT Calibration, and DeepAR
    Aug 20, 2026 #forecasting

    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.
  7. 07
    Irregular Time in Tick Models: Continuous-Time Encodings vs. Plain Positional Embeddings
    Aug 22, 2026 #HFT

    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.
  8. 08
    Adakah Konteks Sepenuh Hari Mengatasi Sepuluh Minit? Perhatian Kilat dan Soalan Panjang Urutan
    Aug 4, 2026 #deep-learning

    Adakah Konteks Sepenuh Hari Mengatasi Sepuluh Minit? Perhatian Kilat dan Soalan Panjang Urutan

    Flash Attention menjadikan konteks dagangan 23,400 langkah bebas pengiraan — jubin, softmax dalam talian dan sempadan IO N^2 d^2 / M. Sama ada konteks yang lebih panjang itu menjadikan model itu lebih baik ialah soalan yang berasingan, dan ia perlu diukur, bukan ditegaskan.
  9. 09
    Knowledge Distillation: Compressing Trading Models for Low-Latency Deployment
    Aug 11, 2026 #model-compression

    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.
  10. 10
    Model Pruning for Low-Latency Trading Inference
    Aug 14, 2026 #model-compression

    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.
  11. 11
    Proses Gaussian untuk Pemodelan Harga Bukan Parametrik
    Aug 6, 2026 #bayesian

    Proses Gaussian untuk Pemodelan Harga Bukan Parametrik

    Reka bentuk kernel untuk siri masa kewangan — Kekasaran ibu, komposisi berkala setempat, campuran spektrum — dan kemungkinan kecil sebagai penyusun tetap yang tidak memerlukan set pengesahan. Selain itu senarai jujur ​​tentang perkara yang masih perlu diukur sebelum mana-mana daripadanya boleh didagangkan.
  12. 12
    Epistemik vs Aleatorik: Mengukur Apa yang Model Pulangan Tidak Tahu
    Jul 31, 2026 #bayesian

    Epistemik vs Aleatorik: Mengukur Apa yang Model Pulangan Tidak Tahu

    Setiap peraturan saiz kedudukan dalam blog ini meruntuhkan ketidakpastian kepada satu skalar. MC Dropout dan ansambl dalam membahagikannya kepada kejahilan model dan hingar pasaran — dan kedua-duanya memerlukan saiz kedudukan yang berbeza.
  13. 13
    Neural ODEs: Does Continuous Time Beat a Delta-Time Feature?
    Aug 16, 2026 #deep-learning

    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.
  14. 14
    Hamiltonian Neural Networks: Does a Financial System Conserve Anything?
    Aug 8, 2026 #deep-learning

    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.
  15. 15
    Operator Neural Fourier untuk Pemodelan Kewangan Berasaskan PDE
    Aug 5, 2026 #deep-learning

    Operator Neural Fourier untuk Pemodelan Kewangan Berasaskan PDE

    Pembelajaran operator memetakan keseluruhan ruang fungsi, bukan titik — cara Operator Neural Fourier meparameterkan operator penyelesaian PDE dalam ruang frekuensi, perkara yang dibeli untuk penentuan harga opsyen, dan tuntutannya yang mana masih perlu diukur pada perkakasan sebenar.