Search Results for author:
Found 6 papers, 2 papers with code
Accelerating Recommender Model Training by Dynamically Skipping Stale Embeddings
Training recommendation models pose significant challenges regarding resource utilization and performance.
Keyformer: KV Cache Reduction through Key Tokens Selection for Efficient Generative Inference
This approach effectively reduces both the KV cache size and memory bandwidth usage without compromising model accuracy.
Ad-Rec: Advanced Feature Interactions to Address Covariate-Shifts in Recommendation Networks
However, deep learning-based recommendation models often face challenges due to evolving user behaviour and item features, leading to covariate shifts.
Integration of Data Driven Technologies in Smart Grids for Resilient and Sustainable Smart Cities: A Comprehensive Review
In addition, the impact of disasters on the power system infrastructure is investigated and different types of optimization techniques that can be used to sustain the power flow in the network during disturbances are compared and analyzed.
Heterogeneous Acceleration Pipeline for Recommendation System Training
This approach utilizes CPU main memory for non-popular embeddings and GPUs' HBM for popular embeddings.
Accelerating Recommendation System Training by Leveraging Popular Choices
This paper leverages this asymmetrical access pattern to offer a framework, called FAE, and proposes a hot-embedding aware data layout for training recommender models.
