jgrusewski 2bdb55cc5b fix(rl): loosen loss defense — penalty 0.01→0.001, stop-loss 2.0→5.0 ...

The previous settings overcorrected: penalty=0.01 + threshold=2.0
taught Q that all trading = punishment → model stopped trading
entirely (wr=1.0, dones=0, hold=91%).

Loosened: penalty=0.001 (10× smaller per-step signal), threshold=5.0
(5× initial risk before force-close). This preserves the loss defense
gradient while giving trades room to develop.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

2026-05-27 22:35:59 +02:00

..

backtesting

…

common

…

config

…

ctrader-openapi

…

data

fix(loader): soft-validate FrontMonth symbol when no streaming SymbolMapping

2026-05-22 18:08:36 +02:00

database

…

market-data

…

ml

fix(ml): migrate test uploads from clone_htod to mapped-pinned

2026-05-22 20:39:32 +02:00

ml-alpha

fix(rl): loosen loss defense — penalty 0.01→0.001, stop-loss 2.0→5.0

2026-05-27 22:35:59 +02:00

ml-asset-selection

…

ml-backtesting

feat(rl): four-layer loss defense + perf sync removal

2026-05-27 22:06:59 +02:00

ml-checkpoint

…

ml-core

fix(ml-core): deterministic GPU weight init via scoped_init_seed

2026-05-19 01:19:13 +02:00

ml-data

…

ml-data-validation

…

ml-dqn

feat(sp22-vnext): Phase B4b-2 — replay-buffer scatter + trainer setter

2026-05-14 02:13:05 +02:00

ml-ensemble

…

ml-explainability

…

ml-features

fix(ml-features): imbalance bar cache on persistent PVC, not /tmp

2026-05-24 23:11:03 +02:00

ml-hyperopt

…

ml-labeling

…

ml-observability

…

ml-paper-trading

…

ml-ppo

…

ml-regime

…

ml-regime-detection

…

ml-risk

…

ml-security

…

ml-stress-testing

…

ml-supervised

…

ml-universe

…

ml-validation

…

model_loader

…

risk

…

risk-data

…

storage

…

trading_engine

…

trading-data

…

training_uploader

…