Search Schedule »

Show All »

Timezone: US/Central »

Indicates the beginning of a session

 Mon Mar 02, 2020

Time Ballroom A
06:30 PM (Posters)
A System for Massively Parallel Hyperparameter Tuning Ballroom A #1
PLink: Discovering and Exploiting Locality for Accelerated Distributed Training on the public Cloud Ballroom A #2
Federated Optimization in Heterogeneous Networks Ballroom A #3
BPPSA: Scaling Back-propagation by Parallel Scan Algorithm Ballroom A #4
Distributed Hierarchical GPU Parameter Server for Massive Scale Deep Learning Ads Systems Ballroom A #5
Resource Elasticity in Distributed Deep Learning Ballroom A #6
SLIDE : In Defense of Smart Algorithms over Hardware Acceleration for Large-Scale Deep Learning Systems Ballroom A #7
FLEET: Flexible Efficient Ensemble Training for Heterogeneous Deep Neural Networks Ballroom A #8
Breaking the Memory Wall with Optimal Tensor Rematerialization Ballroom A #9
What is the State of Neural Network Pruning? Ballroom A #10
SkyNet: a Hardware-Efficient Method for Object Detection and Tracking on Embedded Systems Ballroom A #11
MNN: A Universal and Efficient Inference Engine Ballroom A #12
Willump: A Statistically-Aware End-to-end Optimizer for Machine Learning Inference Ballroom A #13
Attention-based Learning for Missing Data Imputation in HoloClean Ballroom A #14
Privacy-Preserving Bandits Ballroom A #15
Understanding the Downstream Instability of Word Embeddings Ballroom A #16
Model Assertions for Monitoring and Improving ML Models Ballroom A #17
AutoPhase: Juggling HLS Phase Orderings in Random Forests with Deep Reinforcement Learning Ballroom A #18
Automatically batching control-intensive programs for modern accelerators Ballroom A #19
Predictive Precompute with Recurrent Neural Networks Ballroom A #20
Sense & Sensitivities: The Path to General-Purpose Algorithmic Differentiation Ballroom A #21
Ordering Chaos: Memory-Aware Scheduling of Irregularly Wired Neural Networks for Edge Devices Ballroom A #22
Fine-Grained GPU Sharing Primitives for Deep Learning Applications Ballroom A #23
Improving the Accuracy, Scalability, and Performance of Graph Neural Networks with Roc Ballroom A #24
OPTIMUS: OPTImized matrix MUltiplication Structure for Transformer neural network accelerator Ballroom A #25
PoET-BiN: Power Efficient Tiny Binary Neurons Ballroom A #26
Memory-Driven Mixed Low Precision Quantization for Enabling Deep Network Inference on Microcontrollers Ballroom A #27
Trained Quantization Thresholds for Accurate and Efficient Fixed-Point Inference of Deep Neural Networks Ballroom A #28
Riptide: Fast End-to-End Binarized Neural Networks Ballroom A #29
Searching for Winograd-aware Quantized Networks Ballroom A #30
Blink: Fast and Generic Collectives for Distributed ML Ballroom A #31
A Systematic Methodology for Analysis of Deep Learning Hardware and Software Platforms Ballroom A #32
MotherNets: Rapid Deep Ensemble Learning Ballroom A #33
MLPerf Training Benchmark Ballroom A #34