0xReisearch
  • //Welcome
  • //notions
    • Intro
  • //Core
    • Evolution
  • Catalog
    • //hanabi-1
  • //Factory & Core API/SDK (Cross-Framework)
    • Reigent Factory
      • Factory V 0.4.0 Overview
    • How to get your API key
    • API reference
      • GET Reigent
      • Chat Completion
    • ReiCore SDK
      • Integration with Existing Agents
      • Integration with Existing Services
    • Additional Capability (Provided & Custom)
      • DeFi (Current Level : 1)
      • Research
      • Custom Tools
  • Tokenomics
  • API/SDK v0.5 Model - A Base Layer For all Agents
  • /legacy
Powered by GitBook
On this page
  • Mathematical Foundations: Functions and Formulas
  • Training Dynamics and Balanced Validation
  • Model Architecture Details
  • Prediction Temporal Distribution
  • Performance and Future Directions
  • Conclusion
  1. Catalog

//hanabi-1

Hanabi-1 is the the first in our "Catalog" series, which specializes in Financial Prediction. Catalog will be a series of transformer models designed to serve different specialized purposes.

PreviousCatalogNext//Factory & Core API/SDK (Cross-Framework)

Last updated 1 month ago

While the industry gravitates toward increasingly large models, our research has revealed that financial market prediction benefits from a more specialized, compact architecture. Hanabi-1 demonstrates how targeted design can outperform brute-force approaches in specific domains like financial time series analysis

With 16.4 million parameter model consists of:

  • 8 transformer layers with multi-head attention mechanisms

  • 384-dimensional hidden states throughout the network

  • Multiple specialized predictive pathways for direction, volatility, price change, and spread

  • Batch normalization rather than layer normalization for better training dynamics

  • Focal loss implementation to address inherent class imbalance

The compact size enables faster inference times and allows us to deploy models at the edge for real-time decision making—critical for high-frequency market environments.

Mathematical Foundations: Functions and Formulas

Positional Encoding

To help the transformer understand sequence ordering, we implement sinusoidal positional encoding:

PE(pos,2i)=sin⁡(pos⋅1100002i/dmodel)PE(pos,2i)=sin⁡(pos⋅1100002i/dmodel)PE(pos,2i)=sin⁡(pos⋅1100002i/dmodel)

PE(pos,2i+1)=cos⁡(pos⋅1100002i/dmodel)PE(pos,2i+1)=cos⁡(pos⋅1100002i/dmodel)PE(pos,2i+1)=cos⁡(pos⋅1100002i/dmodel)

Where $pos$ is the position within the sequence and $i$ is the dimension index.

Focal Loss for Direction Prediction

To address the severe class imbalance in financial market direction prediction, we implemented Focal Loss:

Where $p_t$ is the model's estimated probability for the correct class and $\gamma$ is the focusing parameter (set to 2.0 in Hanabi-1). This loss function down-weights the contribution of easy examples, allowing the model to focus on harder cases.

Confidence Calibration

A key innovation in Hanabi-1 is our confidence-aware prediction system:

Where $p$ is the predicted probability and $threshold$ is our calibrated decision boundary (0.5). This allows users to filter predictions based on confidence levels, dramatically improving accuracy in high-confidence scenario.

As shown above, predictions with "High" confidence achieve nearly 100% accuracy, while "Very Low" confidence predictions are barely above random chance.

Training Dynamics and Balanced Validation

Training financial models presents unique challenges, particularly the tendency to collapse toward predicting a single class. Our novel validation scoring function addresses this:

Where $PR_{balance}$ is the precision-recall balance metric:

And $Balance_{penalty}$ applies severe penalties for extreme prediction distributions:

if precision == 0 or recall == 0:
    # Heavy penalty for predicting all one class
    balance_penalty = 0.5
elif precision < 0.2 or recall < 0.2:
    # Moderate penalty for extreme imbalance
    balance_penalty = 0.3

This scoring function drives the model toward balanced predictions that maintain high accuracy:

The plot above reveals how training progresses through multiple phases, with early fluctuations stabilizing into consistent improvements after epoch 80.

Model Architecture Details

Hanabi-1 employs a specialized architecture with several innovative components:

  • Feature differentiation through multiple temporal aggregations:

    • Last hidden state capture (most recent information)

    • Average pooling across the sequence (baseline signal)

    • Attention-weighted aggregation (focused signal)

  • Direction pathway with BatchNorm for stable training:

    • Three fully-connected layers with BatchNorm1d

    • LeakyReLU activation (slope 0.1) to prevent dead neurons

    • Xavier initialization with small random bias terms

  • Specialized regression pathways:

    • Separate networks for volatility, price change, and spread prediction

    • Reduced complexity compared to the direction pathway

    • Independent optimization focuses training capacity where needed

The model's multi-task design forces the transformer encoder to learn robust representations that generalize across prediction tasks.

Prediction Temporal Distribution

The distribution of predictions over time shows Hanabi-1's ability to generate balanced directional signals across varying market conditions. Green dots represent correct predictions, and red dots are incorrect predictions.

Performance and Future Directions

Current performance metrics:

  • Direction accuracy: 73.9%

  • F1 score: 0.67

  • Balanced predictions: 54.2% positive / 45.8% negative

Hanabi-1 currently operates on two primary configurations:

  • 4-hour window model (w4_h1)

  • 12-hour window model (w12_h1)

Both predict market movements for the next hour, with the 12-hour window model showing superior performance in more volatile conditions.

Future developments include:

  • Extending prediction horizons to 4, 12 and 24 hours

  • Implementing adaptive thresholds based on market volatility

  • Adding meta-learning approaches for hyperparameter optimization

  • Integrating on-chain signals for cross-domain pattern recognition

Conclusion

Hanabi-1 demonstrates that specialized, compact transformers can achieve remarkable results in financial prediction tasks. By focusing on addressing the unique challenges of financial data—class imbalance, temporal dynamics, and confidence calibration—we've created a model that delivers reliable signals even in challenging market conditions.

While the model can still be refined, we found that it’s a robust and important first step towards the definition and creation of even more capable financial models.

Follow the github repo for the current implementation and future upgrades:

FL(pt)=−(1−pt)γlog⁡(pt)FL(pt)=−(1−pt)γlog⁡(pt)FL(pt)=−(1−pt)γlog⁡(pt)

Confidence=2⋅∣p−threshold∣Confidence=2⋅∣p−threshold∣Confidence=2⋅∣p−threshold∣

Confidence vs Accuracy

ValScore=F1+0.5⋅Accuracy+0.5⋅PRbalance−0.1⋅Loss−BalancepenaltyValScore=F1+0.5⋅Accuracy+0.5⋅PRbalance−0.1⋅Loss−BalancepenaltyValScore=F1+0.5⋅Accuracy+0.5⋅PRbalance−0.1⋅Loss−Balancepenalty

PRbalance=min⁡(Precision,Recall)max⁡(Precision,Recall)PRbalance=min⁡(Precision,Recall)max⁡(Precision,Recall)PRbalance=min⁡(Precision,Recall)max⁡(Precision,Recall)

Training Dynamics
Direction Probabilities

Direction Probabilities
Training Dynamics
Confidence vs Accuracy