Add exposure autoencoder pretraining

README.md

@@ -58,3 +58,15 @@ disease event token + pre-onset exposure embedding -> same next-token Transforme
 ```
 
 The key constraint is that a disease event's own pre-onset exposure must not be used to predict that same disease event.
+
+Pretrain the exposure encoder as a denoising autoencoder using training-set EIDs:
+
+```bash
+python train_exposure_autoencoder.py \
+  --exposure_cache_dir ukb_exposure_cache \
+  --train_eid_file ukb_train_eid.csv
+```
+
+The best checkpoint contains both `model_state_dict`, an `encoder_state_dict`
+compatible with the default gated `TimesNetExposureEncoder`, and the channel
+normalization statistics needed when the encoder is attached to DeepHealth.