Secure FL: Zero-Knowledge Federated Learning

A dual-verifiable framework for federated learning using zero-knowledge proofs to ensure training integrity and aggregation correctness.

Core Features

• Dual ZKP Verification: Client-side zk-STARKs (PySNARK) + Server-side zk-SNARKs (Circom/SnarkJS)
• FedJSCM Aggregation: Momentum-based federated optimization with auto-reinit on architecture changes
• Dynamic Proof Rigor: StabilityMonitor adapts proof complexity (low/medium/high/adaptive) based on training stability
• Deterministic Hashing: SHA-256 over parameter bytes — safe across Python/numpy versions
• Attack Detection: Norm-bound enforcement, hash verification, and ZKP circuit checks block poisoned updates

Architecture

Client: Local Training → zk-STARK proof (learnable params) → Send full state_dict + proof
Server: Receive updates → verify proofs → FedJSCM aggregation → zk-SNARK proof → Distribute model

Dual verification:

1. Clients generate zk-STARK proofs of correct local training (delta norm bound, hash commitment)
2. Server generates zk-SNARK proofs of correct FedJSCM aggregation

Quick Start

Installation

# From source with uv (recommended)
git clone https://github.com/krishantt/secure-fl
cd secure-fl
uv sync --all-extras

ZKP Prerequisites

# Automated setup (installs Rust, Circom, SnarkJS)
make setup-zkp

# Verify
uv run secure-fl check-zkp

Run the attack defense demo

# Synthetic data — fastest, no download needed
uv run python demo/simulate_attack.py --dataset synthetic_small --num-rounds 6

# MNIST
uv run python demo/simulate_attack.py --dataset mnist

# CIFAR-10 with more malicious clients
uv run python demo/simulate_attack.py --dataset cifar10 --malicious 2 --proof-rigor low

# MedMNIST
uv run python demo/simulate_attack.py --dataset medmnist --malicious 2 --proof-rigor low

The TUI shows real-time ZKP verification, attack detection, and accuracy comparison (with ZKP vs projected without). Attackers appear in ~65% of rounds; the rest are all-honest.

Start a federated learning session

# Server
uv run secure-fl-server --config experiments/config.yaml

# Client (in separate terminals)
uv run secure-fl-client --server localhost:8080 --dataset mnist --client-id client_1
uv run secure-fl-client --server localhost:8080 --dataset mnist --client-id client_2

# Or use Docker
docker compose up -d

Python API

from secure_fl.federation.client_runtime import create_client
from secure_fl.models.factory import model_fn

# Create a secure FL client
factory = model_fn("mnist")   # returns MNISTCNN factory
client = create_client(
    client_id="client_1",
    model_fn=factory,
    train_data=train_dataset,
    enable_zkp=True,
    proof_rigor="low",    # "low" | "medium" | "high"
    local_epochs=1,
    learning_rate=0.01,
)

# client.fit(parameters, config) → (updated_params, num_examples, metrics)
# metrics["zkp_proof"] contains the JSON proof string when enable_zkp=True

from secure_fl.federation.aggregation import FedJSCMAggregator
from secure_fl.zkp.proof_manager import ServerProofManager

aggregator = FedJSCMAggregator(momentum=0.9, learning_rate=1.0)
server_pm  = ServerProofManager()

# Verify a client proof
ok = server_pm.verify_client_proof(proof_json, updated_params, global_params)

# Aggregate verified updates
global_params = aggregator.aggregate(
    client_updates=verified_updates,
    client_weights=normalized_weights,
    server_round=rnd,
    global_params=global_params,
)

Configuration

# experiments/config.yaml
server:
  host: "localhost"
  port: 8080
  num_rounds: 10

strategy:
  min_fit_clients: 2
  fraction_fit: 1.0
  momentum: 0.9

zkp:
  enable_zkp: true
  proof_rigor: "high"   # "low" | "medium" | "high" | "adaptive"
  quantize_weights: true
  quantization_bits: 8

Models and Datasets

from secure_fl.models.factory import model_fn
factory = model_fn("cifar10")   # returns callable → ResNet18Model instance
model = factory()

from secure_fl.data.benchmark_data import load_dataset, partition_dataset, subsample_dataset

train, test = load_dataset("mnist", seed=42)
train = subsample_dataset(train, max_samples=2000, seed=42)
client_subsets = partition_dataset(train, n=6, iid=False, seed=42)  # Dirichlet non-IID

Technical Details

FedJSCM Aggregation

m^{t+1} = γ·m^t + (1−γ)·Δ
w^{t+1} = w^t + η·m^{t+1}

where Δ = weighted_avg(client_updates) − global_params. Momentum is reinitialized automatically if the model architecture changes between rounds. Integer-dtype buffers (e.g. BatchNorm num_batches_tracked) are preserved through aggregation.

ZKP Proof Schema

Each client proof is a JSON object containing:

• initial_hash, updated_hash, delta_hash — SHA-256 commitments over learnable parameters
• delta_norm_l2, max_delta_norm_l2 — norm bound enforcement
• pysnark — PySNARK circuit result (delta_bound_proof)

Server verification order: proof present → valid JSON → hash matches → delta hash matches → norm within bound → circuit ok.

Parameter Representations

Two lists exist and must not be mixed:

fit() returns full state dict. ZKP proofs cover learnable-only params. Mixing them with zip causes shape-mismatch errors on BN models (ResNet18, MNISTCNN).

Dynamic Proof Rigor

StabilityMonitor tracks gradient variance and convergence across rounds and adapts ProofRigor:

• High variance / early rounds → high (more proof coverage)
• Stable training → low (faster proofs)
• adaptive mode enables automatic adjustment each round

Experiments & Benchmarks

# Fast smoke run
uv run python experiments/canonical_benchmark.py --quick

# Full reproducible benchmark with repeats
uv run python experiments/canonical_benchmark.py \
  --datasets mnist synthetic_small \
  --num-repeats 5 \
  --seed 42 \
  --require-real-proofs

# Interactive benchmark TUI (supports reattach)
uv run python experiments/benchmark_tui.py

Outputs: benchmark_results.json (per-seed runs) and tradeoff_report.json (aggregated trade-off summary).

Repository Structure

secure-fl/
├── src/secure_fl/
│   ├── core/            # Types, config (thread-safe singleton), exceptions
│   ├── federation/      # SecureFlowerClient, SecureFlowerServer, FedJSCMAggregator,
│   │                    #   StabilityMonitor, client_utils, helpers
│   ├── zkp/             # ClientProofManager, ServerProofManager, quantization
│   ├── models/          # SimpleModel, MNISTCNN, ResNet18Model, factory
│   ├── data/            # benchmark_data: load/partition/subsample datasets
│   ├── cli/             # Click CLI: secure-fl, secure-fl-server, secure-fl-client
│   ├── proofs/
│   │   ├── client_circuits/  # PySNARK delta_bound_proof circuit
│   │   └── server/           # Circom aggregation SNARK circuits
│   └── utils/           # helpers (compute_hash, compute_parameter_norm), logging
├── demo/
│   └── simulate_attack.py    # Textual TUI: real ZKP attack detection demo
├── experiments/
│   ├── canonical_benchmark.py
│   ├── benchmark_tui.py
│   └── config.yaml
└── tests/
    ├── unit/            # No external dependencies (fast)
    └── integration/     # Requires Circom + SnarkJS

Development

make dev          # Full setup (uv sync + ZKP tools)
make lint         # Ruff lint
make format       # Ruff format  ← run before every commit
make type-check   # mypy (strict)
make test         # Full suite
make test-quick   # Fast, --exitfirst
make test-cov     # With coverage (threshold: 34%)

Known Limitations

• _submit_to_blockchain() raises NotImplementedError — not yet implemented
• Server ZKP (Circom/SnarkJS) requires make setup-zkp; subprocess calls time out after 300 s
• Coverage threshold is 34% — being raised incrementally

License

MIT License — see LICENSE for details.

Citation

@misc{timilsina2026securefl,
  title={Secure-FL: Zero-Knowledge Proofs for Dual-Verifiable Federated Learning},
  author={Timilsina, Krishant and Paudel, Bindu and Timilsina, Arun Kumar},
  year={2026},
  url={https://github.com/krishantt/secure-fl},
  note={Preprint}
}

Acknowledgments

• Flower — federated learning infrastructure
• Circom / SnarkJS — zk-SNARK toolchain
• PySNARK — Python zk-STARK/SNARK library