Skip to content

rishabh23002/NumerAi_Meta

Repository files navigation

Meta Learning for Numerai

Project Overview

This project explores a meta-learning approach to the Numerai prediction problem. By treating each era as a unique task, this method allows our models to adapt dynamically to changes in data distribution across different eras.

Modeling Approaches

Deep Learning Models

  • LSTM Model
    • Number of layers: 4
  • Transformer Model
    • Encoder only
    • Number of layers: 4

Meta-Learning Techniques

  1. Model-Agnostic Meta-Learning (MAML)

    • Models trained: MLP, Transformer, LSTM using MAML technique.
  2. MAML + Features from Other Models

    • Combined features from LSTM and Transformer models, totaling 44 features to train an LSTM model using MAML.
  3. MAML with Test-Time Adaptation

    • Employed test-time adaptation techniques to fine-tune models using cosine similarity measures between training and test distributions.
    • Ensemble model was used, integrating outputs from multiple models (Transformer, LSTM, AutoEncoder, MLP, Temporal Block) through a weighted sum computed via MAML.

Ensemble and Adaptation Strategy

  • Ensemble Model: Predictions were generated from five different models and combined using a weighted sum.
  • Weights Generation: A separate model was used to predict weights for combining model outputs based on the training/test feature set.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published