qukan 0.1.2

Official implementation for 'Quantum Kolmogorov Arnold Network' (QuKAN) implementated using PennyLane and PyTorch. Paper: Werner, Y., Malemath, A., Liu, M., Fortes Rey, V., Palaiodimopoulos, N., Lukowicz, P., & Kiefer-Emmanouilidis, M. (2025). QuKAN: A Quantum Circuit Born Machine Approach to Quantum Kolmogorov Arnold Networks. Scientific Reports, 15(1), 35239.

QuKAN: Quantum Kolmogorov Arnold Network

QuKAN is a Python package for Quantum Kolmogorov Arnold Networks, built on top of PennyLane and PyTorch. It inlcudes hybrid and fully quantum neuron architecture implementations.

Features

• Quantum Spline implementation
• Quantum KAN Neurons
• Scalable QuKAN architecture
• Support for QCBM-based spline pretraining

Installation

# Using poetry
poetry install

Usage

import torch
import torch.nn as nn
from qukan import QuKAN

# 1. Initialize model
model = QuKAN(feature_dim=2, num_hlayers=1,strat='QCBM')

# 2. Setup training components
optimizer = torch.optim.Adam(model.parameters(), lr=0.01)
criterion = nn.MSELoss()

# 3. Dummy data
x = torch.rand(10, 2)
y = torch.rand(10, 1).to(torch.float64)

# 4. Training loop
for epoch in range(5):
    optimizer.zero_grad()
    output = model(x)
    loss = criterion(output, y)
    loss.backward()
    optimizer.step()
    print(f"Epoch {epoch+1}, Loss: {loss.item():.4f}")

License

LGPL-3.0-only