Dataset Viewer

The dataset viewer should be available soon. Please retry later.

๐Ÿงฎ COINjecture NP Solutions v5

The Empirical Evolution

Live solutions to NP-hard computational problems from the COINjecture blockchain

License: MIT Dataset Network

Documentation โ€ข GitHub โ€ข Website


๐Ÿ“‹ Overview

This dataset contains real-time solutions to NP-hard problems mined from the COINjecture Network B blockchain. Unlike traditional Proof-of-Work that wastes computational energy, COINjecture implements Proof-of-Useful-Work (PoUW) โ€” every hash contributes to solving computationally significant problems.

What Makes v5 Special

Feature v4 (Previous) v5 (Current)
Tokenomics Hardcoded constants 100% empirical (network-derived)
Light Clients Basic SPV FlyClient + MMR proofs
Node Classification Single type 6 specialized types
Mobile Support None WASM + C-FFI SDK
Metrics Static Live network oracle

๐ŸŽฏ Problem Types

Subset Sum

Given a set of integers, find a subset that sums to a target value.

  • Complexity: NP-complete
  • Applications: Cryptography, resource allocation, financial modeling

Boolean Satisfiability (SAT)

Determine if a Boolean formula can be satisfied.

  • Complexity: NP-complete (Cook-Levin theorem)
  • Applications: Hardware verification, AI planning, scheduling

Traveling Salesman Problem (TSP)

Find the shortest route visiting all cities exactly once.

  • Complexity: NP-hard
  • Applications: Logistics, circuit design, DNA sequencing

Custom Problems

User-submitted computational challenges with bounties.

  • Complexity: Variable (verified NP-hard)
  • Applications: Research, optimization, real-world problems

๐Ÿ“Š Dataset Schema

{
  "problem_id": "uuid-v4",
  "problem_type": "SubsetSum | SAT3 | TSP | Custom",
  "problem_data": {
    "elements": [1, 2, 3, ...],
    "target": 42
  },
  "solution_data": {
    "selected_indices": [0, 2, 5],
    "selected_elements": [1, 3, 10]
  },
  
  "block_height": 12345,
  "timestamp": 1733500000,
  "block_hash": "0x1a2b3c...",
  "prev_block_hash": "0x9f8e7d...",
  
  "work_score": 100.0,
  "solution_quality": 1.0,
  "problem_complexity": 3.5,
  "bounty": "1000000",
  
  "solve_time_us": 150000,
  "verify_time_us": 1200,
  "energy_ratio": 1920.5,
  
  "solver": "12D3KooW...",
  "submitter": "12D3KooW...",
  
  "network_metrics": {
    "hash_rate": 1.5,
    "peer_count": 25,
    "consensus_agreement": 0.95
  }
}

Field Descriptions

Field Type Description
problem_id string Unique identifier for the problem
problem_type enum Category of NP-hard problem
problem_data object Problem-specific input data
solution_data object Verified solution
block_height integer Block number in the chain
timestamp integer Unix timestamp of block creation
work_score float Computational work performed
solution_quality float Optimality measure (1.0 = optimal)
problem_complexity float Estimated problem difficulty
bounty string Reward in microCOIN (u128 as string)
solve_time_us integer Solution time in microseconds
verify_time_us integer Verification time in microseconds
energy_ratio float Verification/solve energy ratio

๐Ÿš€ Quick Start

Load with Hugging Face Datasets

from datasets import load_dataset

# Load the full dataset
dataset = load_dataset("COINjecture/v5")

# Iterate through solutions
for record in dataset["train"]:
    print(f"Block {record['block_height']}: {record['problem_type']}")
    print(f"  Work Score: {record['work_score']}")
    print(f"  Energy Ratio: {record.get('energy_ratio', 'N/A')}x")

Filter by Problem Type

# Get only SAT problems
sat_problems = dataset["train"].filter(
    lambda x: x["problem_type"] == "SAT3"
)

# Get only high-complexity problems
hard_problems = dataset["train"].filter(
    lambda x: x["problem_complexity"] > 4.0
)

Stream Large Datasets

# Stream without downloading entire dataset
dataset = load_dataset("COINjecture/v5", streaming=True)

for record in dataset["train"]:
    process(record)

๐Ÿ“ˆ Statistics

Metric Value
Update Frequency Real-time (every block)
Avg Block Time ~30 seconds
Problem Types 4
Verification Rate 100%
Energy Asymmetry >1000x (v5 with ADZDB)

๐Ÿ”ฌ Research Applications

Machine Learning

  • Train models to predict problem difficulty
  • Learn heuristics for NP-hard optimization
  • Benchmark solver algorithms

Cryptography

  • Study hash function distributions
  • Analyze computational hardness assumptions
  • Research post-quantum implications

Distributed Systems

  • Study consensus mechanisms
  • Analyze network behavior under load
  • Research incentive-compatible protocols

๐Ÿงฎ Empirical Tokenomics (v5)

v5 introduces zero hardcoded constants. All economic parameters are derived from live network state:

โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”
โ”‚                    NETWORK METRICS ORACLE                       โ”‚
โ”œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ค
โ”‚                                                                 โ”‚
โ”‚  Hash Rate    โ”€โ”€โ†’  Emission Bounds                             โ”‚
โ”‚  Solve Times  โ”€โ”€โ†’  Problem Hardness Factor                     โ”‚
โ”‚  Median Fees  โ”€โ”€โ†’  Base Storage Cost                           โ”‚
โ”‚  Stake Dist   โ”€โ”€โ†’  Staking Thresholds                          โ”‚
โ”‚  Fault Impact โ”€โ”€โ†’  Reputation Severities                       โ”‚
โ”‚                                                                 โ”‚
โ”‚  Formula: value = f(network_state)                             โ”‚
โ”‚  Result:  Self-regulating, governance-free economics           โ”‚
โ”‚                                                                 โ”‚
โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜

๐Ÿ“ฑ LightSync Protocol

v5 implements FlyClient with Merkle Mountain Ranges for O(log n) chain verification:

Protocol Bandwidth Use Case
Full Sync O(n) Archive nodes
SPV O(n) headers Desktop wallets
FlyClient O(log n) Mobile devices

For 1M blocks:

  • SPV: ~80 MB of headers
  • FlyClient: ~50 KB (proofs + sampled headers)

๐Ÿ”— Related Resources


๐Ÿ“œ Citation

@dataset{coinjecture_v5_2025,
  title     = {COINjecture NP Solutions Dataset v5},
  author    = {COINjecture Network Contributors},
  year      = {2024},
  publisher = {Hugging Face},
  url       = {https://huggingface.co/datasets/COINjecture/v5},
  note      = {Real-time NP-hard problem solutions from Proof-of-Useful-Work blockchain}
}

โš–๏ธ License

This dataset is released under the MIT License. You are free to use, modify, and distribute the data for any purpose, including commercial applications.


Built with ๐Ÿง  by the COINjecture community

Where every hash solves something meaningful

Downloads last month
3,516