WideDeep Tutorial: Wide & Deep Learning

WideDeep Tutorial: Wide & Deep Learning#

Sandbox Model — Experimental

This model is part of the CoreRec Sandbox — an experimental collection of algorithms for research and exploration. Sandbox models are not production-tested and may have incomplete implementations. For production-ready models, see the Production Models.

Introduction#

WideDeep jointly trains wide linear models for memorization and deep neural networks for generalization.

Paper: Cheng et al. 2016 - Wide & Deep Learning for Recommender Systems (Google)

How WideDeep Works#

Architecture#

Hybrid Architecture:

  1. Wide Component: Generalized Linear Model (GLM)

    • Memorizes co-occurrence of features

    • Uses cross-product transformations

  2. Deep Component: Feed-Forward Neural Network

    • Generalizes to unseen feature combinations

    • Uses low-dimensional embeddings

  3. Joint Training: Weighted sum of both components

Key Innovation: Combining benefits of memorization (Wide) and generalization (Deep)

Architecture:

Wide Part (Linear)      Deep Part (DNN)
      ↓                       ↓
    Output <----- Sigmoid(Sum)

Mathematical Foundation#

Prediction:

P(y=1|x) = σ(w_wide^T [x, φ(x)] + w_deep^T a^(lf) + b)

where:

  • φ(x): Cross-product transformations (Wide)

  • a^(lf): Output of last deep layer

  • σ: Sigmoid function

Joint Optimization: Backpropagate gradients to both parts simultaneously using FTRL (Wide) and AdaGrad (Deep).

Reference Implementation#

Implementation Note

The code examples below are reference implementations for learning purposes. This is a sandbox model and has not been production-tested. Verify behavior thoroughly before using in production.

Step 1: Import and Load Data#

from corerec.sandbox.collaborative_full.nn_base.WideDeep_base import WideDeep_base as WideDeep
from cr_learn import ml_1m
from sklearn.model_selection import train_test_split
import numpy as np

# Load dataset with CORRECT API
data = ml_1m.load()  # Returns dict with 'ratings', 'users', 'movies'
ratings_df = data['ratings']  # DataFrame with user_id, movie_id, rating, timestamp

print(f"Loaded {len(ratings_df)} ratings")

# Split data
train_df, test_df = train_test_split(ratings_df, test_size=0.2, random_state=42)

# Extract arrays for model
train_users = train_df['user_id'].values
train_items = train_df['movie_id'].values
train_ratings = train_df['rating'].values

test_users = test_df['user_id'].values
test_items = test_df['movie_id'].values
test_ratings = test_df['rating'].values

Step 2: Initialize Model#

model = WideDeep(
    name="WideDeep_Model",
    wide_features=['genre', 'os'],
    deep_features=['age', 'install_history'],
    hidden_units=[1024, 512, 256],
    epochs=20,
    batch_size=256,
    learning_rate=0.001,
    verbose=True
)

print(f"Initialized {model.name}")

Step 3: Train#

model.fit(
    user_ids=train_users,
    item_ids=train_items,
    ratings=train_ratings
)

print("Training complete!")

Step 4: Predict#

# Single prediction
score = model.predict(user_id=1, item_id=100)
print(f"Predicted score: {score:.3f}")

# Batch predictions
test_predictions = model.batch_predict(list(zip(test_users[:100], test_items[:100])))

Step 5: Recommend#

# Get top-10 recommendations for user
user_id = 1
recommendations = model.recommend(
    user_id=user_id,
    top_k=10
)

print(f"Top-10 recommendations for User {user_id}:")
for rank, item_id in enumerate(recommendations, 1):
    print(f"  {rank}. Item {item_id}")

Step 6: Evaluate#

from sklearn.metrics import mean_squared_error
import numpy as np

# Predict all test ratings
test_pred = [model.predict(u, i) for u, i in zip(test_users, test_items)]
rmse = np.sqrt(mean_squared_error(test_ratings, test_pred))
print(f"Test RMSE: {rmse:.4f}")

Step 7: Save & Load#

# Save model
model.save('artifacts/widedeep_model')

# Load model
loaded = WideDeep.load('artifacts/widedeep_model')
test_score = loaded.predict(1, 100)
print(f"Loaded model prediction: {test_score:.3f}")

Key Takeaways#

When to Use WideDeep#

✅ Perfect For:

  • App stores (Google Play)

  • Recommending new & old items

  • Large-scale regression/classification

  • Mixed feature types (sparse + dense)

  • Production systems needing low latency

❌ Not For:

  • Pure collaborative filtering (no features)

  • Sequential patterns (use RNN/Transformer)

  • Small datasets (overfitting risk)

Best Practices#

  1. Wide Features: Cross-product of important categorical features

  2. Deep Features: Embeddings for sparse, raw values for dense

  3. Optimizers: FTRL for Wide (sparsity), Adam/AdaGrad for Deep

  4. Hidden Layers: [1024, 512, 256] typical

  5. Embedding Dim: 32-128

  6. Batch Size: Large (thousands)

Scaling & Production Considerations#

If you plan to move this sandbox model toward production use, consider:

  • Data Volume: Test with realistic dataset sizes; sandbox implementations may not be optimized for large-scale data.

  • Distributed Training: For datasets exceeding single-machine memory, consider wrapping the model in distributed training frameworks (e.g., PyTorch DDP, Horovod).

  • Hyperparameter Tuning: Use systematic search (Optuna, Ray Tune) rather than manual tuning. The defaults in sandbox models are not tuned for any specific domain.

  • Serving Latency: Profile inference time. Some architectures (e.g., attention-based) may need quantization or ONNX export for low-latency serving.

  • Monitoring: Implement A/B testing and online metrics (CTR, engagement, diversity) before replacing an existing system.

  • Validation: Write comprehensive unit tests and integration tests before deploying. Sandbox models do not have CI coverage.

Further Reading#

  • Paper: Cheng et al. 2016 - Wide & Deep Learning for Recommender Systems (Google)

  • GitHub Repository

Contents