[신간의 별자리x] 우리/미술, 그리고 ‘슬픔의 박물관’

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Data Mining - Concepts and Techniques.

Large language models have been shown to achieve remarkable performance across a variety of natural language tasks using few-shot learning , which drastically reduces the number of task-speciﬁc training examples needed to adapt the model to a particular application. To further our understanding of the impact of scale on few-shot learning, we trained a 540-billion parameter, densely activated, Transformer language model, which we call Pathways Language Model (PaLM). We trained PaLM on 6144 TPU v4 chips using Pathways, a new ML system which enables highly eﬃcient training across multiple TPU Pods. We demonstrate continued beneﬁts of scaling by achieving state-of-the-art few-shot learning results on hundreds of language understanding and generation benchmarks. On a number of these tasks, PaLM 540B achieves breakthrough performance, outperforming the ﬁnetuned state-of-the-art on a suite of multi-step reasoning tasks, and outperforming average human performance on the recently released BIG-bench benchmark. A signiﬁcant number of BIG-bench tasks showed discontinuous improvements from model scale, meaning that performance steeply increased as we scaled to our largest model. PaLM also has strong capabilities in multilingual tasks and source code generation, which we demonstrate on a wide array of benchmarks. We additionally provide a comprehensive analysis on bias and toxicity, and study the extent of training data memorization with respect to model scale. Finally, we discuss the ethical considerations related to large language models and discuss potential mitigation strategies.

PaLM: Scaling Language Modeling with Pathways

We explore how generating a chain of thought -- a series of intermediate reasoning steps -- significantly improves the ability of large language models to perform complex reasoning. In particular, we show how such reasoning abilities emerge naturally in sufficiently large language models via a simple method called chain of thought prompting, where a few chain of thought demonstrations are provided as exemplars in prompting. Experiments on three large language models show that chain of thought prompting improves performance on a range of arithmetic, commonsense, and symbolic reasoning tasks. The empirical gains can be striking. For instance, prompting a 540B-parameter language model with just eight chain of thought exemplars achieves state of the art accuracy on the GSM8K benchmark of math word problems, surpassing even finetuned GPT-3 with a verifier.

Chain of Thought Prompting Elicits Reasoning in Large Language Models

Detection of execution anomalies is very important for the maintenance, development, and performance refinement of large scale distributed systems. Execution anomalies include both work flow errors and low performance problems. People often use system logs produced by distributed systems for troubleshooting and problem diagnosis. However, manually inspecting system logs to detect anomalies is unfeasible due to the increasing scale and complexity of distributed systems. Therefore, there is a great demand for automatic anomalies detection techniques based on log analysis. In this paper, we propose an unstructured log analysis technique for anomalies detection. In the technique, we propose a novel algorithm to convert free form text messages in log files to log keys without heavily relying on application specific knowledge. The log keys correspond to the log-print statements in the source code which can provide cues of system execution behavior. After converting log messages to log keys, we learn a Finite State Automaton (FSA) from training log sequences to present the normal work flow for each system component. At the same time, a performance measurement model is learned to characterize the normal execution performance based on the log mes-sages’ timing information. With these learned models, we can automatically detect anomalies in newly input log files. Experiments on Hadoop and SILK show that the technique can effectively detect running anomalies.

/pdf/execution-anomaly-detection-in-distributed-systems-through-k62fowugdy.pdf

Execution Anomaly Detection in Distributed Systems through Unstructured Log Analysis

Detecting execution anomalies is very important to the maintenance and monitoring of large-scale distributed systems. People often use console logs that are produced by distributed systems for troubleshooting and problem diagnosis. However, manually inspecting console logs for the detection of anomalies is unfeasible due to the increasing scale and complexity of distributed systems. Therefore, there is great demand for automatic anomaly detection techniques based on log analysis. In this paper, we propose an unstructured log analysis technique for anomaly detection, with a novel algorithm to automatically discover program invariants in logs. At first, a log parser is used to convert the unstructured logs to structured logs. Then, the structured log messages are further grouped to log message groups according to the relationship among log parameters. After that, the program invariants are automatically mined from the log message groups. The mined invariants can reveal the inherent linear characteristics of program work flows. With these learned invariants, our technique can automatically detect anomalies in logs. Experiments on Hadoop show that the technique can effectively detect execution anomalies. Compared with the state of art, our approach can not only detect numerous real problems with high accuracy but also provide intuitive insight into the problems.

/pdf/mining-invariants-from-console-logs-for-system-problem-3l65porm2b.pdf

Mining invariants from console logs for system problem detection

System logs are widely used in various tasks of software system management. It is crucial to avoid logging too little or too much. To achieve so, developers need to make informed decisions on where to log and what to log in their logging practices during development. However, there exists no work on studying such logging practices in industry or helping developers make informed decisions. To fill this significant gap, in this paper, we systematically study the logging practices of developers in industry, with focus on where developers log. We obtain six valuable findings by conducting source code analysis on two large industrial systems (2.5M and 10.4M LOC, respectively) at Microsoft. We further validate these findings via a questionnaire survey with 54 experienced developers in Microsoft. In addition, our study demonstrates the high accuracy of up to 90% F-Score in predicting where to log.

/pdf/where-do-developers-log-an-empirical-study-on-logging-36s14uh91c.pdf

Where do developers log? an empirical study on logging practices in industry

Logging is a common programming practice of practical importance to collect system runtime information for postmortem analysis. Strategic logging placement is desired to cover necessary runtime information without incurring unintended consequences (e.g., Performance overhead, trivial logs). However, in current practice, there is a lack of rigorous specifications for developers to govern their logging behaviours. Logging has become an important yet tough decision which mostly depends on the domain knowledge of developers. To reduce the effort on making logging decisions, in this paper, we propose a "learning to log" framework, which aims to provide informative guidance on logging during development. As a proof of concept, we provide the design and implementation of a logging suggestion tool, Log Advisor, which automatically learns the common logging practices on where to log from existing logging instances and further leverages them for actionable suggestions to developers. Specifically, we identify the important factors for determining where to log and extract them as structural features, textual features, and syntactic features. Then, by applying machine learning techniques (e.g., Feature selection and classifier learning) and noise handling techniques, we achieve high accuracy of logging suggestions. We evaluate Log Advisor on two industrial software systems from Microsoft and two open-source software systems from Git Hub (totally 19.1M LOC and 100.6K logging statements). The encouraging experimental results, as well as a user study, demonstrate the feasibility and effectiveness of our logging suggestion tool. We believe our work can serve as an important first step towards the goal of "learning to log".

/pdf/learning-to-log-helping-developers-make-informed-logging-4udvidm9uh.pdf

Learning to log: helping developers make informed logging decisions

Successful software maintenance is becoming increasingly critical due to the increasing dependence of our society and economy on software systems. One key problem of software maintenance is the difficulty in understanding the evolving software systems. Program workflows can help system operators and administrators to understand system behaviors and verify system executions so as to greatly facilitate system maintenance. In this paper, we propose an algorithm to automatically discover program workflows from event traces that record system events during system execution. Different from existing workflow mining algorithms, our approach can construct concurrent workflows from traces of interleaved events. Our workflow mining approach is a three-step coarse-to-fine algorithm. At first, we mine temporal dependencies for each pair of events. Then, based on the mined pair-wise tem-poral dependencies, we construct a basic workflow model by a breadth-first path pruning algorithm. After that, we refine the workflow by verifying it with all training event traces. The re-finement algorithm tries to find out a workflow that can interpret all event traces with minimal state transitions and threads. The results of both simulation data and real program data show that our algorithm is highly effective.

/pdf/mining-program-workflow-from-interleaved-traces-5fi0bzky0a.pdf

Qiang Fu

Papers

Execution Anomaly Detection in Distributed Systems through Unstructured Log Analysis

Mining invariants from console logs for system problem detection

Where do developers log? an empirical study on logging practices in industry

Learning to log: helping developers make informed logging decisions

Mining program workflow from interleaved traces