tzervas/bwsk-efficientnet-b0

BWSK EfficientNet-B0

EfficientNet-B0 (5M params) trained in 6 variants (3 BWSK modes x 2 experiments) on CIFAR-10 with full convergence training and early stopping.

This repo contains all model weights, configs, and training results in a single consolidated repository.

What is BWSK?

BWSK is a framework that classifies every neural network operation as S-type (information-preserving, reversible, coordination-free) or K-type (information-erasing, synchronization point) using combinator logic. This classification enables reversible backpropagation through S-phases to save memory, and CALM-based parallelism analysis.

Model Overview

Property	Value
Base Model	google/efficientnet-b0
Architecture	Cnn (image_cls)
Parameters	5M
Dataset	CIFAR-10
Eval Metric	Accuracy

S/K Classification

Type	Ratio
S-type (information-preserving)	33.5%
K-type (information-erasing)	59.6%
Gray (context-dependent)	7.0%

Fine-tune Results

Mode	Final Loss	Val Accuracy	Test Accuracy	Peak Memory	Time	Epochs
Conventional	0.3806	89.0%	89.6%	2.8 GB	57s	2
BWSK Analyzed	0.2952	89.3%	88.5%	2.8 GB	58s	2
BWSK Reversible	0.2530	90.1%	90.0%	2.8 GB	57s	2

Memory savings (reversible vs conventional): 0.0%

From Scratch Results

Mode	Final Loss	Val Accuracy	Test Accuracy	Peak Memory	Time	Epochs
Conventional	0.2993	87.4%	87.4%	2.8 GB	9.0m	10
BWSK Analyzed	0.5080	79.4%	78.8%	2.8 GB	4.8m	6
BWSK Reversible	0.3454	88.1%	87.1%	2.8 GB	8.9m	10

Memory savings (reversible vs conventional): 0.0%

Repository Structure

├── README.md
├── results.json
├── finetune-conventional/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json
├── finetune-bwsk-analyzed/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json
├── finetune-bwsk-reversible/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json
├── scratch-conventional/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json
├── scratch-bwsk-analyzed/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json
├── scratch-bwsk-reversible/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json

Usage

Load a specific variant:

import torch
# Load fine-tuned conventional variant
# Weights are in the finetune-conventional/ subdirectory

Training Configuration

Setting	Value
Optimizer	AdamW
LR (fine-tune)	1e-03
LR (from-scratch)	5e-03
LR Schedule	Cosine with warmup
Max Grad Norm	1.0
Mixed Precision	AMP (float16)
Early Stopping	Patience 3
Batch Size	32

Links

• GitHub Repository
• Whitepaper
• Full Training Report

Citation

@software{zervas2026bwsk,
  author = {Zervas, Tyler},
  title = {BWSK: Combinator-Typed Neural Network Analysis},
  year = {2026},
  url = {https://github.com/tzervas/ai-s-combinator},
}

License

MIT