Augmented Negative Sampling for Collaborative Filtering logo
AI Tool Profile

Augmented Negative Sampling for Collaborative Filtering

To balance efficiency and effectiveness, the vast majority of existing methods follow the two-pass approach, in which the first pass samples a fixed number of unobserved items by a simple static distribution and then the second pass selects the final negative items using a more sophisticated negative sampling strategy.

Website
github.com
Pricing model
Free
Price start
Free

GitHub Link

The GitHub link is https://github.com/asa9aotk/ans-recbole

Introduce

The GitHub repository "Asa9aoTK/ANS-RecBole" contains an implementation of ANS using the RecBole framework. The ANS method can be found in the directory "recbole/model/general_recommender." The repository includes four distinct base models, and among them, LightGCN is offered as a base model version for ease of access. The code for BPR-MF, SGL, and NGCF is also present, made accessible to readers through straightforward adjustments. To balance efficiency and effectiveness, the vast majority of existing methods follow the two-pass approach, in which the first pass samples a fixed number of unobserved items by a simple static distribution and then the second pass selects the final negative items using a more sophisticated negative sampling strategy.

Content

We implemented ANS using RecBole framework, please find our method under recbole/model/general_recommender. We have used four different base models in this paper, and we have provided LightGCN as a version of the base model for your convenience. The code for BPR-MF, SGL, and NGCF is available to the reader with simple modifications.

Alternatives & Similar Tools

LongLLaMA-handle very long text contexts, up to 256,000 tokens logo

LongLLaMA is a large language model designed to handle very long text contexts, up to 256,000 tokens. It's based on OpenLLaMA and uses a technique called Focused Transformer (FoT) for training. The repository provides a smaller 3B version of LongLLaMA for free use. It can also be used as a replacement for LLaMA models with shorter contexts.

LAMA: Human motion data to realistic complex 3D model actions logo

LAMA utilizes a reinforcement learning framework combined with a motion matching algorithm. Reinforcement learning helps the model make appropriate decisions in various scenarios, while motion matching algorithms ensure that synthesized actions match real human actions. In addition, LAMA also utilizes the motion editing framework of manifold learning to cover various possible changes in interactions and operations.

Compare Augmented Negative Sampling for Collaborative Filtering

Quick compare routes for nearby alternatives.

All compare routes →