Search Results for author:
Found 32 papers, 13 papers with code
Reparo: Loss-Resilient Generative Codec for Video Conferencing
Packet loss during video conferencing often leads to poor quality and video freezing.
Counterfactual Identifiability of Bijective Causal Models
We study counterfactual identifiability in causal models with bijective generation mechanisms (BGM), a class that generalizes several widely-used causal models in the literature.
Online Reinforcement Learning in Non-Stationary Context-Driven Environments
We present Locally Constrained Policy Optimization (LCPO), an online RL approach that combats CF by anchoring policy outputs on old experiences while optimizing the return on current experiences.
FactorJoin: A New Cardinality Estimation Framework for Join Queries
Neither classical nor learning-based methods yield satisfactory performance when estimating the cardinality of the join queries.
Gemino: Practical and Robust Neural Compression for Video Conferencing
We design Gemino, a new neural compression system for video conferencing based on a novel high-frequency-conditional super-resolution pipeline.
Demystifying Reinforcement Learning in Time-Varying Systems
Such agents must explore and learn new environments, without hurting the system's performance, and remember them over time.
CausalSim: A Causal Framework for Unbiased Trace-Driven Simulation
Key to CausalSim is mapping unbiased trace-driven simulation to a tensor completion problem with extremely sparse observations.
Efficient Strong Scaling Through Burst Parallel Training
As emerging deep neural network (DNN) models continue to grow in size, using large GPU clusters to train DNNs is becoming an essential requirement to achieving acceptable training times.
Updating Street Maps using Changes Detected in Satellite Imagery
To address this challenge, much work has studied automatically processing geospatial data sources such as GPS trajectories and satellite images to reduce the cost of maintaining digital maps.
Efficient Video Compression via Content-Adaptive Super-Resolution
SRVC decodes the video by passing the decompressed low-resolution video frames through the (time-varying) super-resolution model to reconstruct high-resolution video frames.
TagMe: GPS-Assisted Automatic Object Annotation in Videos
We show TagMe can produce high-quality object annotations in a fully-automatic and low-cost way.
Inferring High-Resolution Traffic Accident Risk Maps Based on Satellite Imagery and GPS Trajectories
Traffic accidents cost about 3% of the world's GDP and are the leading cause of death in children and young adults.
Cortex: Harnessing Correlations to Boost Query Performance
Unlike prior work, Cortex can adapt itself to any existing primary index, whether single or multi-dimensional, to harness a broad variety of correlations, such as those that exist between more than two attributes or have a large number of outliers.
Neural Large Neighborhood Search
Results on several datasets show that it is possible to learn a neighbor selection policy that allows LNS to efficiently find good solutions.
Real-world Video Adaptation with Reinforcement Learning
Client-side video players employ adaptive bitrate (ABR) algorithms to optimize user quality of experience (QoE).
Sat2Graph: Road Graph Extraction through Graph-Tensor Encoding
Inferring road graphs from satellite imagery is a challenging computer vision task.
Tsunami: A Learned Multi-dimensional Index for Correlated Data and Skewed Workloads
Filtering data based on predicates is one of the most fundamental operations for any modern data warehouse.
Real-Time Video Inference on Edge Devices via Adaptive Model Streaming
Real-time video inference on edge devices like mobile phones and drones is challenging due to the high computation cost of Deep Neural Networks.
RoadTagger: Robust Road Attribute Inference with Graph Neural Networks
The usage of graph neural networks allows information propagation on the road network graph and eliminates the receptive field limitation of image classifiers.
Learning Multi-dimensional Indexes
Scanning and filtering over multi-dimensional tables are key operations in modern analytical database engines.
Learning Generalizable Device Placement Algorithms for Distributed Machine Learning
We present Placeto, a reinforcement learning (RL) approach to efficiently find device placements for distributed neural network training.
Inferring and Improving Street Maps with Data-Driven Automation
Through an evaluation on a large-scale dataset including satellite imagery, GPS trajectories, and ground-truth map data in forty cities, we show that Mapster makes automation practical for map editing, and enables the curation of map datasets that are more complete and up-to-date at less cost.
Practical Low Latency Proof of Work Consensus
Bitcoin is the first fully-decentralized permissionless blockchain protocol to achieve a high level of security, but at the expense of poor throughput and latency.
Distributed, Parallel, and Cluster Computing Cryptography and Security Networking and Internet Architecture
Placeto: Learning Generalizable Device Placement Algorithms for Distributed Machine Learning
Unlike prior approaches that only find a device placement for a specific computation graph, Placeto can learn generalizable device placement policies that can be applied to any graph.
Machine-Assisted Map Editing
Systems to automatically infer road network graphs from aerial imagery and GPS trajectories have been proposed to improve coverage of road maps.
Adaptive Neural Signal Detection for Massive MIMO
We propose MMNet, a deep learning MIMO detection scheme that significantly outperforms existing approaches on realistic channels with the same or lower computational complexity.
Understanding & Generalizing AlphaGo Zero
AlphaGo Zero (AGZ) introduced a new {\em tabula rasa} reinforcement learning algorithm that has achieved superhuman performance in the games of Go, Chess, and Shogi with no prior knowledge other than the rules of the game.
Learning Scheduling Algorithms for Data Processing Clusters
Efficiently scheduling data processing jobs on distributed compute clusters requires complex algorithms.
Restructuring Endpoint Congestion Control
Each datapath—such as the Linux kernel TCP, UDP-based QUIC, or kernel-bypass transports like mTCP-on-DPDK—summarizes information about packet round-trip times, receptions, losses, and ECN via a well-defined interface to algorithms running in the off-datapath Congestion Control Plane (CCP).
Variance Reduction for Reinforcement Learning in Input-Driven Environments
We consider reinforcement learning in input-driven environments, where an exogenous, stochastic input process affects the dynamics of the system.
Graph2Seq: Scalable Learning Dynamics for Graphs
By not limiting the representation to a fixed dimension, Graph2Seq scales naturally to graphs of arbitrary sizes and shapes.
RoadTracer: Automatic Extraction of Road Networks from Aerial Images
Mapping road networks is currently both expensive and labor-intensive.
