• Publications
  • Influence
pFabric: minimal near-optimal datacenter transport
TLDR
We present pFabric, a minimalistic datacenter transport design that provides near theoretically optimal flow completion times even at the 99th percentile for short flows, while still minimizing average flow completion time for long flows. Expand
  • 528
  • 96
  • PDF
SecondNet: a data center network virtualization architecture with bandwidth guarantees
TLDR
In this paper, we propose virtual data center (VDC) as the unit of resource allocation for multiple tenants in the cloud. Expand
  • 600
  • 64
  • PDF
Deconstructing datacenter packet transport
TLDR
We present, pFabric, a minimalistic datacenter fabric design that provides near-optimal performance in terms of completion time for high-priority flows and overall network utilization. Expand
  • 63
  • 6
  • PDF
LRW-1000: A Naturally-Distributed Large-Scale Benchmark for Lip Reading in the Wild
TLDR
We present a naturally-distributed large-scale benchmark for lip-reading in the wild, named LRW-1000, which contains 1,000 classes with 718,018 samples from more than 2,000 individual speakers. Expand
  • 28
  • 6
  • PDF
Signet Ring Cell Detection with a Semi-supervised Learning Framework
TLDR
We propose a semi-supervised learning framework for the signet ring cell detection problem. Expand
  • 19
  • 6
  • PDF
Neural network ensembles: combining multiple models for enhanced performance using a multistage approach
TLDR
In this paper, some methods for creating ensembles are reviewed, including the following approaches: methods of selecting diverse training data from the original source data set, constructing different neural network models, selecting ensemble nets from ensemble candidates and combining ensemble members' results. Expand
  • 59
  • 5
PCube: Improving Power Efficiency in Data Center Networks
TLDR
We present PCube, a server-centric data center structure that conserves energy by varying bandwidth availability based on traffic demand. Expand
  • 46
  • 5
  • PDF
Tree-structured Data Regeneration in Distributed Storage Systems with Regenerating Codes
TLDR
We propose a new design, referred to as RCTREE, that combines the advantage of regenerating codes with a tree-structured regeneration topology. Expand
  • 88
  • 4
  • PDF
Multi-task Sparse Learning with Beta Process Prior for Action Recognition
TLDR
In this paper, we formulate human action recognition as a novel Multi-Task Sparse Learning(MTSL) framework which aims to construct a test sample with multiple features from as few bases as possible. Expand
  • 40
  • 3
  • PDF
Personalized Modeling of Facial Action Unit Intensity
TLDR
In this paper, we propose a two-step approach for personalized modeling of facial AU intensity from spontaneously displayed facial expressions. Expand
  • 17
  • 3
  • PDF