EDEN-AlexNet-Custom-ImageNet300 — Baseline

Primary KPI: EAG (Energy-to-Accuracy Gradient) = -9.1826e-11 ΔAcc/ΔJoules

Abstract

This model is part of Project EDEN (Energy-Driven Evolution of Networks), implementing the E2AM (Energy Efficient Advanced Model) Framework. The goal is to shift AI benchmarking from pure accuracy to Green SOTA — maximising predictive power per Joule consumed.

Applied Technique: Baseline – Standard Full Training (Reference Study)

Profiling Environment

Component	Specification
GPU	NVIDIA GeForce GTX 1080 Ti (11 GB VRAM, 250 W TDP)
CPU	Intel Xeon W-2125 (4 cores / 8 threads @ 4.00 GHz)
RAM	63.66 GB System RAM
OS	Windows 10
Dataset	Custom-ImageNet300 — ~450,000 images – 300 classes (224 px)

🟢 Green Delta Table

Comparing this model against the reference baseline (ResNet-50 equivalent)

Metric	ResNet50 Baseline	AlexNet (EDEN)	Δ
Accuracy	0.9573	0.9861	+2.88%
Total Energy (J)	380,392,115	66,625,795	82.48% saved
CO₂ Emissions (kg)	50.1906	8.7909	—
EAG Score	—	-9.1826e-11	ΔAcc/ΔJoules

A positive EAG means this model learns more per Joule than the baseline. A negative EAG indicates a trade-off where higher accuracy required more energy investment.

E2AM Algorithm — Applied Phases

Standard full fine-tuning used as the Brute-Force Baseline for energy comparison. All layers trained from epoch 1 with a fixed learning rate and no gradient accumulation. Included for transparent EAG benchmarking.

Training Statistics

Metric	Value
Final Accuracy	0.9861 (98.61%)
Total Energy Consumed	66,625,795 J (18.5072 kWh)
Training Time	4,081 s (1.13 hrs)
Estimated CO₂	8.7909 kg CO₂e
Training Log	test2\alexnet_CustomImageNet300_stats.csv

📊 Training Visualizations

Accuracy & Energy over Training

Green = accuracy (left axis) · Orange dashed = cumulative energy (right axis)

EAG Metric Trajectory

EAG = ΔAccuracy / ΔJoules — positive means learning more per Joule than baseline

Project-Wide Overview

All EDEN models: energy vs accuracy

Cite This Research

@misc{eden2025,
  title     = {Project EDEN: Energy-Driven Evolution of Networks},
  author    = {EDEN Research Team},
  year      = {2025},
  note      = {Hugging Face: Shanmuk4622},
  url       = {https://huggingface.co/Shanmuk4622}
}