Search Results for author:
Found 7 papers, 2 papers with code
Saturn: Efficient Multi-Large-Model Deep Learning
In this paper, we propose Saturn, a new data system to improve the efficiency of multi-large-model training (e. g., during model selection/hyperparameter optimization).
Saturn: An Optimized Data System for Large Model Deep Learning Workloads
Such models need multiple GPUs due to both their size and computational load, driving the development of a bevy of "model parallelism" techniques & tools.
InTune: Reinforcement Learning-based Data Pipeline Optimization for Deep Recommendation Models
Our studies lead us to design and build a new solution for data pipeline optimization, InTune.
Systems for Parallel and Distributed Large-Model Deep Learning Training
Deep learning (DL) has transformed applications in a variety of domains, including computer vision, natural language processing, and tabular data analysis.
Hydra: A System for Large Multi-Model Deep Learning
In this paper, we present Hydra, a system designed to tackle such challenges by enabling out-of-the-box scaling for multi-large-model DL workloads on even commodity GPUs in a resource-efficient manner.
Ranked #5 on
Language Modelling
on WikiText-2
(using extra training data)
Model-Parallel Model Selection for Deep Learning Systems
As deep learning becomes more expensive, both in terms of time and compute, inefficiencies in machine learning (ML) training prevent practical usage of state-of-the-art models for most users.
Gradient-Based Algorithms for Machine Teaching
This is formulated as a problem of functional optimization where, at each teaching iteration, the teacher seeks to align the steepest descent directions of the risk of (1) the teaching set and (2) entire example population.
