Publications (22)
SPECTRA: Synthetic IR Test Collections with Relevance Oracles and Controlled Distractor Diagnostics
Eric Liang
Scalable information retrieval testing needs corpora that are large enough to stress index construction, ranking latency, query routing, and evaluation tooling, yet human-judged te…
Tune: A Research Platform for Distributed Model Selection and Training
Richard Liaw, Eric Liang, Robert Nishihara +3
Modern machine learning algorithms are increasingly computationally demanding, requiring specialized hardware and distributed computation to achieve high performance in a reasonabl…
Anycast Performance in Context
Eric Liang
IP anycast lets a service advertise one address from many physical sites, leaving BGP to map each client to a site. It is central to the DNS root server system, public resolvers, a…
Deep Unsupervised Cardinality Estimation
Zongheng Yang, Eric Liang, Amog Kamsetty +7
Cardinality estimation has long been grounded in statistical tools for density estimation. To capture the rich multivariate distributions of relational tables, we propose the use o…
Ray: A Distributed Framework for Emerging AI Applications
Philipp Moritz, Robert Nishihara, Stephanie Wang +8
The next generation of AI applications will continuously interact with the environment and learn from these interactions. These applications impose new and demanding systems requir…
Predicting COVID-19 Prevalence Using Wastewater RNA Surveillance: A Semi-Supervised Learning Approach with Temporal Feature Trust
Yifei Chen, Eric Liang
As COVID-19 transitions into an endemic disease that remains constantly present in the population at a stable level, monitoring its prevalence without invasive measures becomes inc…
The Streaming Batch Model for Efficient and Fault-Tolerant Heterogeneous Execution
Frank Sifei Luan, Ron Yifeng Wang, Yile Gu +12
While ML model training and inference are both GPU-intensive, CPU-based data processing is often the bottleneck. Distributed data processing systems based on the batch or stream pr…
Exoshuffle-CloudSort
Frank Sifei Luan, Stephanie Wang, Samyukta Yagati +7
We present Exoshuffle-CloudSort, a sorting application running on top of Ray using the Exoshuffle architecture. Exoshuffle-CloudSort runs on Amazon EC2, with input and output data…
Acceptance-Test-Driven Evaluation Protocols for Business-Centric LLM Systems
Eric Liang
Large language model (LLM) applications are increasingly expected to satisfy deterministic institutional requirements while relying on probabilistic generative components. This mis…
RLlib Flow: Distributed Reinforcement Learning is a Dataflow Problem
Eric Liang, Zhanghao Wu, Michael Luo +3
Researchers and practitioners in the field of reinforcement learning (RL) frequently leverage parallel computation, which has led to a plethora of new algorithms and systems in the…
SEMBridge: Tagless-Final Program Semantics with Weakest-Precondition and Bounded-Checking Interpretations
Eric Liang
Formal methods provide rigorous accounts of program behavior, but practical software engineering often works through executable libraries, tests, and incremental design. This paper…
Exoshuffle: An Extensible Shuffle Architecture
Frank Sifei Luan, Stephanie Wang, Samyukta Yagati +7
Shuffle is one of the most expensive communication primitives in distributed data processing and is difficult to scale. Prior work addresses the scalability challenges of shuffle b…
NeuroCard: One Cardinality Estimator for All Tables
Zongheng Yang, Amog Kamsetty, Sifei Luan +4
Query optimizers rely on accurate cardinality estimates to produce good execution plans. Despite decades of research, existing cardinality estimators are inaccurate for complex que…
Neural Packet Classification
Eric Liang, Hang Zhu, Xin Jin +1
Packet classification is a fundamental problem in computer networking. This problem exposes a hard tradeoff between the computation and state complexity, which makes it particularl…
Predicting Pedestrian Crosswalk Behavior Using Convolutional Neural Networks
Eric Liang, Mark Stamp
A common yet potentially dangerous task is the act of crossing the street. Pedestrian accidents contribute a significant amount to the high number of annual traffic casualties, whi…
Variable Skipping for Autoregressive Range Density Estimation
Eric Liang, Zongheng Yang, Ion Stoica +3
Deep autoregressive models compute point likelihood estimates of individual data points. However, many applications (i.e., database cardinality estimation) require estimating range…
Feature-Optimized Vision for Adaptive 3D Scene Reconstruction
Eric Liang
Three-dimensional scene reconstruction depends on local image evidence that is both visually discriminative and geometrically useful. Fixed feature thresholds and uniform feature b…
Hoplite: Efficient and Fault-Tolerant Collective Communication for Task-Based Distributed Systems
Siyuan Zhuang, Zhuohan Li, Danyang Zhuo +5
Task-based distributed frameworks (e.g., Ray, Dask, Hydro) have become increasingly popular for distributed applications that contain asynchronous and dynamic workloads, including…
IMPACT: Importance Weighted Asynchronous Architectures with Clipped Target Networks
Michael Luo, Jiahao Yao, Richard Liaw +2
The practical usage of reinforcement learning agents is often bottlenecked by the duration of training time. To accelerate training, practitioners often turn to distributed reinfor…
SECUREVENT: Hybrid AI/ML Security Monitoring for Distributed Event-Based Systems
Eric Liang
Distributed event-based systems have become a common substrate for Internet-scale publish/subscribe services, IoT telemetry, cloud-native microservices, and security operations pip…
RLlib: Abstractions for Distributed Reinforcement Learning
Eric Liang, Richard Liaw, Philipp Moritz +6
Reinforcement learning (RL) algorithms involve the deep nesting of highly irregular computation patterns, each of which typically exhibits opportunities for distributed computation…
Population Based Augmentation: Efficient Learning of Augmentation Policy Schedules
Daniel Ho, Eric Liang, Ion Stoica +2
A key challenge in leveraging data augmentation for neural network training is choosing an effective augmentation policy from a large search space of candidate operations. Properly…