Henry Hoffmann

Henry Hoffmann

Auto-generated profile, based on publicly available information

Henry Hoffmann

Department of Computer Science, University of Chicago, Chicago, Illinois, USA

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 August 2024
Verified
Conference Paper
Open access
YLYuhan Liu
Yuhan Liu
HLHanchen Li
Hanchen Li
YCYihua Cheng
Yihua Cheng
SRSiddhant Ray
Siddhant Ray
YHYuyang Huang
Yuyang Huang
JJJunchen Jiang
Junchen Jiang
As large language models (LLMs) take on complex tasks, their inputs are supplemented with longer contexts that incorporate domain knowledge. Yet using long contexts is challenging as nothing can be generated until the whole context is processed by the LLM. While the context-processing delay can be reduced by reusing the KV cache of a context across different inputs, fetching the KV cache, which contains large tensors, over the network can cause high extra network delays. CacheGen is a fast context-loading module for LLM systems. First, CacheGen uses a custom tensor encoder, leveraging KV cache's distributional properties to encode a KV cache into more compact bitstream representations with negligible decoding overhead, to save bandwidth usage. Second, CacheGen adapts the compression level of different parts of a KV cache to cope with changes in available bandwidth, in order to maintain low context-loading delay and high generation quality. We test CacheGen on popular LLMs and datasets. Compared to the recent systems that reuse the KV cache, CacheGen reduces the KV cache size by 3.5--4.3x and the total delay in fetching and processing contexts by 3.2--3.7x with negligible impact on the LLM response quality. Our code is at: https://github.com/UChi-JCL/CacheGen.

11 October 2024
Verified
Conference Paper
Open access
VXVasco Xu
Vasco Xu
CGChenfeng Gao
Chenfeng Gao
HHHenry Hoffmann
Henry Hoffmann
KAKaran Ahuja
Karan Ahuja
There has been a continued trend towards minimizing instrumentation for full-body motion capture, going from specialized rooms and equipment, to arrays of worn sensors and recently sparse inertial pose capture methods. However, as these techniques migrate towards lower-fidelity IMUs on ubiquitous commodity devices, like phones, watches, and earbuds, challenges arise including compromised online performance, temporal consistency, and loss of global translation due to sensor noise and drift. Addressing these challenges, we introduce MobilePoser, a real-time system for full-body pose and global translation estimation using any available subset of IMUs already present in these consumer devices. MobilePoser employs a multi-stage deep neural network for kinematic pose estimation followed by a physics-based motion optimizer, achieving state-of-the-art accuracy while remaining lightweight. We conclude with a series of demonstrative applications to illustrate the unique potential of MobilePoser across a variety of fields, such as health and wellness, gaming, and indoor navigation to name a few.

21 May 2022
Verified
Conference Paper
Open access
CWChengcheng Wan
Chengcheng Wan
SLShicheng Liu
Shicheng Liu
SXSophie Xie
Sophie Xie
YLYifan Liu
Yifan Liu
HHHenry Hoffmann
Henry Hoffmann
SLShan Lu
Shan Lu
An increasing number of software applications incorporate machine learning (ML) solutions for cognitive tasks that statistically mimic human behaviors. To test such software, tremendous human effort is needed to design image/text/audio inputs that are relevant to the software, and to judge whether the software is processing these inputs as most human beings do. Even when misbehavior is exposed, it is often unclear whether the culprit is inside the cognitive ML API or the code using the API. This paper presents Keeper, a new testing tool for software that uses cognitive ML APIs. Keeper designs a pseudo-inverse function for each ML API that reverses the corresponding cognitive task in an empirical way (e.g., an image search engine pseudo-reverses the image-classification API), and incorporates these pseudo-inverse functions into a symbolic execution engine to automatically generate relevant image/text/audio inputs and judge output correctness. Once misbehavior is exposed, Keeper attempts to change how ML APIs are used in software to alleviate the misbehavior. Our evaluation on a variety of open-source applications shows that Keeper greatly improves the branch coverage, while identifying many previously unknown bugs.

27 October 2014 in IEEE Computer Architecture Letters
Verified
Journal Article
Closed access
QSQingchuan Shi
Qingchuan Shi
HHHenry Hoffmann
Henry Hoffmann
OKOmer Khan
Omer Khan
To protect multicores from soft-error perturbations, resiliency schemes have been developed with high coverage but high power/performance overheads (~2x). We observe that not all soft-errors affect program correctness, some soft-errors only affect program accuracy, i.e., the program completes with certain acceptable deviations from soft-error free outcome. Thus, it is practical to improve processor efficiency by trading off resilience overheads with program accuracy. We propose the idea of declarative resilience that selectively applies resilience schemes to both crucial and non-crucial code, while ensuring program correctness. At the application level, crucial and non-crucial code is identified based on its impact on the program outcome. The hardware collaborates with software support to enable efficient resilience with 100 percent soft-error coverage. Only program accuracy is compromised in the worst-case scenario of a soft-error strike during non-crucial code execution. For a set of multithreaded benchmarks, declarative resilience improves completion time by an average of 21 percent over state-of-the-art hardware resilience scheme that protects all executed code. Its performance overhead is ~1.38x over a multicore that does not support resilience.

22 February 2019 in IEEE Software
Verified
Journal Article
Closed access
SBSaeid Barati
Saeid Barati
FBFerenc Bartha
Ferenc Bartha
SBSwarnendu Biswas
Swarnendu Biswas
RCRobert Cartwright
Robert Cartwright
ADAdam Duracz
Adam Duracz
SZSizhuo Zhang
Sizhuo Zhang
Our software framework, Proteus, treats adaptation as a first-class object, enabling rapid development of robust, adaptive applications. Proteus developers specify their programs' intent and adaptable components (or knobs). A control-theoretic runtime continually monitors the running application, adjusting knobs so that the specified intent is met.

14 March 2015 in ACM SIGPLAN Notices
Verified
Journal Article
Open access
NMNikita Mishra
Nikita Mishra
HZHuazhe Zhang
Huazhe Zhang
JLJohn D. Lafferty
John D. Lafferty
HHHenry Hoffmann
Henry Hoffmann
In many deployments, computer systems are underutilized -- meaning that applications have performance requirements that demand less than full system capacity. Ideally, we would take advantage of this under-utilization by allocating system resources so that the performance requirements are met and energy is minimized. This optimization problem is complicated by the fact that the performance and power consumption of various system configurations are often application -- or even input -- dependent. Thus, practically, minimizing energy for a performance constraint requires fast, accurate estimations of application-dependent performance and power tradeoffs. This paper investigates machine learning techniques that enable energy savings by learning Pareto-optimal power and performance tradeoffs. Specifically, we propose LEO, a probabilistic graphical model-based learning system that provides accurate online estimates of an application's power and performance as a function of system configuration. We compare LEO to (1) offline learning, (2) online learning, (3) a heuristic approach, and (4) the true optimal solution. We find that LEO produces the most accurate estimates and near optimal energy savings.