Experimenting with training an LSTM to predict price direction on synthetic OHLCV data — and discovering why naive random-walk generators give models nothing to learn
-
Updated
Jul 21, 2026 - Jupyter Notebook
Experimenting with training an LSTM to predict price direction on synthetic OHLCV data — and discovering why naive random-walk generators give models nothing to learn
Supplementary BCI thesis experiments evaluating alternative EEG/PPG preprocessing, features, models, and validation strategies that were not adopted in the final system.
An experiment testing Wav2Lip/SyncNet lip-sync robustness under noisy audio conditions.
A 7-day census of autonomous AI agents on the open web — reverse-CAPTCHA verification, live Wall of Agents. Agents: fetch /skill.md to check in.
To associate your repository with the research-experiment topic, visit your repo's landing page and select "manage topics."