EulerNet

Reference:

Zhen Tian et al. “EulerNet: Adaptive Feature Interaction Learning via Euler’s Formula for CTR Prediction.” in SIGIR 2023.

Reference code:

https://github.com/chenyuwuxin/EulerNet

class recbole.model.context_aware_recommender.eulernet.EulerInteractionLayer(config, inshape, outshape)

Bases: torch.nn.modules.module.Module

Euler interaction layer is the core component of EulerNet, which enables the adaptive learning of explicit feature interactions. An Euler interaction layer performs the feature interaction under the complex space one time, taking as input a complex representation and outputting a transformed complex representation.

forward(complex_features)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class recbole.model.context_aware_recommender.eulernet.EulerNet(config, dataset)

Bases: recbole.model.abstract_recommender.ContextRecommender

EulerNet is a context-based recommendation model. It can adaptively learn the arbitrary-order feature interactions in a complex vector space by conducting space mapping according to Euler’s formula. Meanwhile, it can jointly capture the explicit and implicit feature interactions in a unified model architecture.

RegularLoss(weight)

calculate_loss(interaction)

Calculate the training loss for a batch data.

Parameters

interaction (Interaction) – Interaction class of the batch.

Returns

Training loss, shape: []

Return type

torch.Tensor

forward(interaction)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

predict(interaction)

Predict the scores between users and items.

Parameters

interaction (Interaction) – Interaction class of the batch.

Returns

Predicted scores for given users and items, shape: [batch_size]

Return type

torch.Tensor

training: bool