Search Results for author:
Found 10 papers, 3 papers with code
Heterogeneous federated collaborative filtering using FAIR: Federated Averaging in Random Subspaces
To include data from all devices in federated learning, we must enable collective training of embedding tables on devices with heterogeneous memory capacities.
In defense of parameter sharing for model-compression
In this paper, we comprehensively assess the trade-off between memory and accuracy across RPS, pruning techniques, and building smaller models.
The trade-offs of model size in large recommendation models : A 10000 $\times$ compressed criteo-tb DLRM model (100 GB parameters to mere 10MB)
It contains 100GB of embedding memory (25+Billion parameters).
Efficient model compression with Random Operation Access Specific Tile (ROAST) hashing
Advancements in deep learning are often associated with increasing model sizes.
Raw Nav-merge Seismic Data to Subsurface Properties with MLP based Multi-Modal Information Unscrambler
This paradigm breaks the SI into local inversion tasks, which predicts each small chunk of subsurface properties using surrounding seismic data.
Embedding models through the lens of Stable Coloring
Grounded on this framework, we show that many algorithms ranging across different domains are, in fact, searching for continuous stable coloring solutions of an underlying graph corresponding to the domain.
Random Offset Block Embedding Array (ROBE) for CriteoTB Benchmark MLPerf DLRM Model : 1000$\times$ Compression and 3.1$\times$ Faster Inference
In this paper, we present Random Offset Block Embedding Array (ROBE) as a low memory alternative to embedding tables which provide orders of magnitude reduction in memory usage while maintaining accuracy and boosting execution speed.
Beyond Convolutions: A Novel Deep Learning Approach for Raw Seismic Data Ingestion
We propose a fundamental shift to move away from convolutions and introduce SESDI: Set Embedding based SDI approach.
Semantically Constrained Memory Allocation (SCMA) for Embedding in Efficient Recommendation Systems
In particular, our LMA embeddings achieve the same performance compared to standard embeddings with a 16$\times$ reduction in memory footprint.
Density Sketches for Sampling and Estimation
We introduce Density sketches (DS): a succinct online summary of the data distribution.
