Files

T

KiHoLee 6c8471ece0 Match figure styling and labels to the submitted manuscript

Enlarge the in-canvas fonts and line weights of the result figures so
that they stay legible at the printed column width, split the Fig. 5
convergence curve into a pre-meta adaptation entry and the proposed
MAML entry, and rename the autoencoder legend to match the table row.
Add the analytic complexity replot behind Fig. 4, which was missing
from the repository, and correct the table numbering in the README.

2026-08-03 12:49:53 +09:00

3.8 KiB

Executable File

Raw Blame History

JSAC_AIRAN

Code and stored results for the manuscript "AI-Native Multi-User Semantic Communications via Meta-Learned Attention for 6G AI-RAN" (submitted to IEEE Journal on Selected Areas in Communications, Special Issue on Towards Open and Intelligent 6G RAN).

This repository contains exactly the code needed to reproduce the experiments reported in the paper, together with the stored CSV results from which every figure is regenerated. All experiments use fixed random seeds.

What each script does

Script	Role in the paper
`c11_doppler_csi.py`	Shared library. Implements the time-varying tapped-delay-line channel by sample-level time-domain convolution, so intra-symbol Doppler produces genuine inter-carrier interference, plus the pilot-aging frame structure and the aging-aware LMMSE front end of Section IV-A.
`c13_mnist.py`	MNIST study of Sections V-A and V-B. Defines the signed user-wise attention receiver, the Transformer shared-embedding separator, and the per-user autoencoder, with joint training, warm-started first-order MAML meta-training, and the SNR-sweep evaluation behind Fig. 6.
`c18_mnist_doppler.py`	Doppler sweep behind Fig. 7, with task-conditional decoder-side adaptation at every operating point.
`c19_mnist_epoch.py`	Convergence study behind Fig. 5, recording the task-adapted SER at every training checkpoint.
`c22_maml_epoch.py`	Budget-matched meta-training trajectory that forms the right-hand segment of Fig. 5.
`c21_mnist_flat.py`	Flat Rayleigh study behind Table II, including the decoder-side first-order MAML variant.
`c20_bert.py`	Concluding BERT/AG News text study behind Fig. 8, from feature caching to training and evaluation.
`fig_replot_complexity.py`	Analytic complexity comparison behind Fig. 4, for the softmax and the signed realizations.

Reproducing the figures

Each figure regenerates from the stored CSV results without retraining.

python3 fig_replot_complexity.py           # Fig. 4
python3 c19_mnist_epoch.py --mode fig      # Fig. 5
python3 c13_mnist.py --mode fig            # Fig. 6
python3 c18_mnist_doppler.py --mode fig    # Fig. 7
python3 c20_bert.py --mode fig             # Fig. 8

Fig. 4 is analytic and needs no stored results. In Fig. 5 the green curve is split into two legend entries: left of the dotted line the five-step adaptation starts from the jointly trained model and no meta-training has taken place, and right of it the receiver meta-trains within the same total step budget.

Table II values are stored in results_mnist/mnist_flat.csv, and Table III values come from results_mnist/mnist_results.csv and results_mnist/mnist_doppler.csv.

Rerunning the experiments

Full reruns retrain from scratch on the same fixed seeds.

# MNIST time-varying study (Figs. 5-7, Table IV)
python3 c13_mnist.py --mode train
python3 c13_mnist.py --mode train-tf
python3 c13_mnist.py --mode train-ae
python3 c13_mnist.py --mode train-tx-cls
python3 c13_mnist.py --mode train-maml
python3 c13_mnist.py --mode eval
python3 c18_mnist_doppler.py --mode run
python3 c19_mnist_epoch.py --mode run
python3 c19_mnist_epoch.py --mode run-maml --steps 7500
python3 c22_maml_epoch.py --pretrain-steps 7500

# Flat Rayleigh study (Table III)
python3 c21_mnist_flat.py --mode run

# BERT text study (Fig. 8)
python3 c20_bert.py --mode cache
python3 c20_bert.py --mode train-tx-cls
python3 c20_bert.py --mode train
python3 c20_bert.py --mode train-tf
python3 c20_bert.py --mode train-ae
python3 c20_bert.py --mode train-maml
python3 c20_bert.py --mode eval

Environment

WSL2 Ubuntu with Python 3 and CUDA-enabled PyTorch. The BERT study additionally needs transformers and datasets. MNIST downloads automatically through torchvision, and AG News through datasets (fancyzhx/ag_news).