| 발표 |
이병철 교수(광주과학기술원)
| 시간 |
16:05-16:30
|
|---|
| 제목 |
Adaptive Correction of Sampling Bias in Dynamic Call Graphs
|
|---|
| 요약 |
This talk introduces a practical low-overhead adaptive technique of correcting sampling bias in profiling dynamic call graphs. Timer-based sampling keeps the overhead low but sampling bias lowers the accuracy when either observable call events or sampling actions are not equally spaced in time. To mitigate sampling bias, our adaptive correction technique weights each sample by monitoring time-varying spacing of call events and sampling actions. We implemented and evaluated our adaptive correction technique in Jikes RVM, a high-performance virtual machine. In our empirical evaluation, our technique significantly improved the sampling accuracy without measurable overhead and resulted in effective feedback directed inlining.
|
|---|
| 발표 |
안기영박사 (Portland State University 졸업)
| 시간 |
16:30-16:55
|
|---|
| 제목 |
Type Inference Prototyping Engines from Relational specifications of type systems (TIPER)
|
|---|
| 요약 |
TIPER aims to automatically generate Type Inference Prototyping Engine from Relational specifications of type systems. Our ultimate goal for the TIPER project is to build a framework that automates type system implementations just as parser implementations are automated by parser generators, ever since Yacc was developed in the 70's. This project will reduce the high-cost of type system implementations supporting type inference so that advancements in type systems research would become much more available to the developers as a cost-effective technology to support in language implementations. This talk will introduce our preliminary results on specifying various degrees of parametric polymorphism found in functional languages using logic programming and briefly introduce the plans and progress regarding the TIPER project -- homepage
|
|---|
| 발표 |
오학주 교수(고려대학교)
| 시간 |
16:55-17:20
|
|---|
| 제목 |
Machine Learning Approaches to Adaptive Program Analysis
|
|---|
| 요약 |
프로그램 분석에서 정확도와 비용사이의 적절한 균형을 머신러닝 기법을 이용해 달성해오고 있는 연구를 소개한다. 정적 프로그램 분석 기술을 효과적으로 사용하기 위해서는 큰 프로그램을 정확하게 분석 할 수 있어야 한다. 이를 위해서는 비용이 큰 정확도 향상 기법들을 대상 프로그램에서 효과적인 부분에만 선별적으로 적용할 수 있는 기술이 필요하다. 이 발표에서는 이 문제를 머신 러닝 기법들을 활용해서 해결할 수 있음을 보이고, 최근 성과와 함께 현재 진행중인 연구 내용들을 소개한다.
|
프로그래밍언어연구회
|---|
