KOINEU Logo
Supervised Random Walks: Predicting and Recommending Links in Social   Networks

Supervised Random Walks: Predicting and Recommending Links in Social Networks

: ์†Œ์…œ ๋„คํŠธ์›Œํฌ์—์„œ ๋งํฌ ์˜ˆ์ธก์€ ๋งค์šฐ ์ค‘์š”ํ•œ ๋ฌธ์ œ์ž…๋‹ˆ๋‹ค. ์ด๋Š” ๋„คํŠธ์›Œํฌ์˜ ์—ญ๋™์„ฑ๊ณผ ๋…ธ๋“œ ๊ฐ„์˜ ์ƒํ˜ธ ์ž‘์šฉ์„ ์ดํ•ดํ•˜๋Š” ๋ฐ ํ•ต์‹ฌ์ ์ธ ์—ญํ• ์„ ํ•˜๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค. ๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ๊ธฐ์กด ์—ฐ๊ตฌ๋“ค์ด ์ „์ฒด ๋„คํŠธ์›Œํฌ ๊ตฌ์กฐ์— ์ดˆ์ ์„ ๋งž์ถ˜ ๋ฐ˜๋ฉด, ๊ฐœ๋ณ„ ์—ฃ์ง€ ์ƒ์„ฑ ์ˆ˜์ค€์—์„œ์˜ ๋ฉ”์ปค๋‹ˆ์ฆ˜์„ ํƒ๊ตฌํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค. ์†Œ์…œ ๋„คํŠธ์›Œํฌ๊ฐ€ ์–ด๋–ป๊ฒŒ ์ง„ํ™”ํ•˜๊ณ  ์ƒˆ๋กœ์šด ๋งํฌ๊ฐ€ ํ˜•์„ฑ๋˜๋Š”์ง€์— ๋Œ€ํ•œ ๊ทผ๋ณธ์ ์ธ ์งˆ๋ฌธ์„ ๋˜์ง€๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์—ฐ๊ตฌ์ง„์€ ๊ฐ๋…ํ˜• ๋žœ๋ค ์›Œํฌ๋ผ๋Š” ๋…ํŠนํ•œ ์ ‘๊ทผ ๋ฐฉ์‹์„ ์ œ์•ˆํ•ฉ๋‹ˆ๋‹ค. ์ด ๋ฐฉ๋ฒ•์€ ๋„คํŠธ์›Œํฌ ๊ตฌ์กฐ์™€ ๋…ธ๋“œ ๋ฐ ์—ฃ์ง€ ์†์„ฑ์„ ๊ฒฐํ•ฉํ•˜์—ฌ ๋ฏธ๋ž˜์˜ ๋งํฌ ์ƒ์„ฑ์— ๋Œ€ํ•œ ํ†ต์ฐฐ๋ ฅ์„ ์ œ๊ณต

Physics Network Data Structures Statistics Social Networks Artificial Intelligence Machine Learning Computer Science
No Image

Surf3R: Rapid Surface Reconstruction from Sparse RGB Views in Seconds

Surf3R์€ 3D ํ‘œ๋ฉด ์žฌ๊ตฌ์„ฑ ๋ถ„์•ผ์—์„œ ์ค‘์š”ํ•œ ์ง„์ „์„ ์ด๋ฃน๋‹ˆ๋‹ค. ๊ธฐ์กด ๋ฐฉ๋ฒ•๊ณผ ๋‹ฌ๋ฆฌ, ์นด๋ฉ”๋ผ ๋ณด์ •์ด๋‚˜ ์ž์„ธ ์ถ”์ •์„ ์š”๊ตฌํ•˜์ง€ ์•Š์œผ๋ฏ€๋กœ ๋ณต์žกํ•˜๊ณ  ์‹œ๊ฐ„ ์†Œ๋ชจ์ ์ธ ์‚ฌ์ „ ์ฒ˜๋ฆฌ๋ฅผ ํ”ผํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋Š” ์‹ค์„ธ๊ณ„ ์• ํ”Œ๋ฆฌ์ผ€์ด์…˜์— ๋Œ€ํ•œ 3D ์žฌ๊ตฌ์ถ•์˜ ์‹ค์šฉ์„ฑ์„ ํฌ๊ฒŒ ํ–ฅ์ƒ์‹œํ‚ต๋‹ˆ๋‹ค. ๋˜ํ•œ ์ œ์•ˆ๋œ ๋‹ค์ค‘ ๋ธŒ๋žœ์น˜ ์•„ํ‚คํ…์ฒ˜๋Š” ๋‹ค์–‘ํ•œ ๋ทฐ์—์„œ ์ •๋ณด๋ฅผ ํšจ๊ณผ์ ์œผ๋กœ ํ†ตํ•ฉํ•˜์—ฌ ์ „์ฒด ์žฅ๋ฉด์„ ์ดํ•ดํ•˜๋Š” ๋ฐ ๋„์›€์ด ๋ฉ๋‹ˆ๋‹ค. D Normal ์ •๊ทœํ™”๊ธฐ๋Š” ํ‘œ๋ฉด ์„ธ๋ถ€ ์‚ฌํ•ญ๊ณผ ์ผ๊ด€์„ฑ์— ๋Œ€ํ•œ ์ •ํ™•๋„๋ฅผ ๋†’์ด๊ณ , ์ด๋Š” ํŠนํžˆ ๋ณต์žกํ•œ ํ™˜๊ฒฝ์ด๋‚˜ ๋…ธ์ด์ฆˆ๊ฐ€ ๋งŽ์€ ๋ฐ์ดํ„ฐ์— ์œ ์šฉํ•ฉ๋‹ˆ๋‹ค. Surf3R์˜ ๋น 

Survey on Individual Differences in Visualization

Survey on Individual Differences in Visualization

: ๋ฐ์ดํ„ฐ ์‹œ๊ฐํ™”๋Š” ๋‹ค์–‘ํ•œ ๋ถ„์•ผ์—์„œ ๋„๋ฆฌ ์‚ฌ์šฉ๋˜๋Š” ๊ฐ•๋ ฅํ•œ ๋„๊ตฌ์ด์ง€๋งŒ, ๋ชจ๋“  ์‚ฌ์šฉ์ž๊ฐ€ ๋™์ผํ•œ ๋ฐฉ์‹์œผ๋กœ ์‹œ๊ฐํ™”๋ฅผ ์ดํ•ดํ•˜๊ณ  ์ƒํ˜ธ ์ž‘์šฉํ•˜๋Š” ๊ฒƒ์€ ์•„๋‹™๋‹ˆ๋‹ค. ๊ฐœ์ธ์˜ ์ฐจ์ด๊ฐ€ ์ด๋Ÿฌํ•œ ๊ณผ์ •์— ๋ฏธ์น˜๋Š” ์˜ํ–ฅ์€ ๋งค์šฐ ์ค‘์š”ํ•˜๋ฉฐ, ์ด๋Š” ์‹œ์Šคํ…œ์˜ ํšจ์œจ์„ฑ๊ณผ ์ ‘๊ทผ์„ฑ์— ์ง์ ‘์ ์ธ ์˜ํ–ฅ์„ ๋ฏธ์น  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์‹ฌ๋ฆฌํ•™์  ์—ฐ๊ตฌ๋Š” ์„ฑ๊ฒฉ๊ณผ ์ธ์ง€ ๋Šฅ๋ ฅ์ด ์ž‘์—… ์ˆ˜ํ–‰์— ๋ฏธ์น˜๋Š” ์˜ํ–ฅ์„ ๋ฐํ˜€๋ƒˆ์Šต๋‹ˆ๋‹ค. ์˜ˆ๋ฅผ ๋“ค์–ด, ์™ธํ–ฅ์ ์ธ ์‚ฌ๋žŒ๋“ค์€ ์ผ๋ฐ˜์ ์œผ๋กœ ๋‚ดํ–ฅ์ ์ธ ์‚ฌ๋žŒ๋“ค๋ณด๋‹ค ๋” ์ ๊ทน์ ์œผ๋กœ ์ฐธ์—ฌํ•˜๊ณ , ์ƒˆ๋กœ์šด ์•„์ด๋””์–ด๋ฅผ ํƒ๊ตฌํ•˜๋Š” ๊ฒฝํ–ฅ์ด ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ์ฐจ์ด๋Š” ๋ฐ์ดํ„ฐ ์‹œ๊ฐํ™” ์‹œ์Šคํ…œ์˜ ์‚ฌ์šฉ ๋ฐฉ์‹์—๋„ ๋ฐ˜

Computer Science HCI
No Image

SymLight: Exploring Interpretable and Deployable Symbolic Policies for Traffic Signal Control

์‹ฌ๋ผ์ดํŠธ(SymLight)๋Š” ๊ตํ†ต ์‹ ํ˜ธ ์ œ์–ด(TSC)๋ฅผ ์œ„ํ•œ ๊ธฐํ˜ธ ์ •์ฑ… ํƒ์ƒ‰ ํ”„๋ ˆ์ž„์›Œํฌ๋กœ์„œ, MCTS ๊ธฐ๋ฐ˜์˜ ์šฐ์„ ์ˆœ์œ„ ํ•จ์ˆ˜ ๊ฒ€์ƒ‰ ๋ฐฉ๋ฒ•์„ ์‚ฌ์šฉํ•ฉ๋‹ˆ๋‹ค. ์ด ์ ‘๊ทผ๋ฒ•์€ ์‹ ๊ฒฝ๋ง ๊ธฐ๋ฐ˜ ์ •์ฑ…๋“ค์˜ ๊ณผ๋„ํ•œ ๋งค๊ฐœ๋ณ€์ˆ˜ํ™”์™€ ํˆฌ๋ช…์„ฑ ๋ถ€์กฑ ๋ฌธ์ œ๋ฅผ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•ด ์„ค๊ณ„๋˜์—ˆ์Šต๋‹ˆ๋‹ค. ์‹ฌ๋ผ์ดํŠธ๋Š” ๊ฐ„๊ฒฐํ•˜๋ฉด์„œ๋„ ํ‘œํ˜„๋ ฅ์ด ๋›ฐ์–ด๋‚œ ์šฐ์„ ์ˆœ์œ„ ํ•จ์ˆ˜ ํ‘œํ˜„๋ฒ•์„ ์ œ์•ˆํ•˜์—ฌ ํ–‰๋™ ๊ณต๊ฐ„์˜ ์กฐํ•ฉ ํญ๋ฐœ ๋ฌธ์ œ๋ฅผ ์™„ํ™”์‹œํ‚ต๋‹ˆ๋‹ค. ๋˜ํ•œ ๊ตฌ์กฐ์  ๋กค์•„์›ƒ ์ „๋žต์„ ํ†ตํ•ด ์ด์ „์— ๋ฐœ๊ฒฌ๋œ ๊ณ ํ’ˆ์งˆ ์šฐ์„ ์ˆœ์œ„ ํ•จ์ˆ˜์˜ ๊ตฌ์กฐ๋ฅผ ํ™œ์šฉํ•จ์œผ๋กœ์จ, ํƒ์ƒ‰ ๊ณผ์ •์„ ํšจ์œจ์ ์œผ๋กœ ์œ ๋„ํ•ฉ๋‹ˆ๋‹ค. ์ด ๋ฐฉ๋ฒ•์€ ์‹ค์„ธ๊ณ„ ๋ฐ์ดํ„ฐ์…‹์—์„œ ๋‹ค์–‘ํ•œ ๋ฒ ์ด

Symmetries for the Ablowitz-Ladik hierarchy: I. Four-potential case

Symmetries for the Ablowitz-Ladik hierarchy: I. Four-potential case

๋ณธ ์—ฐ๊ตฌ๋Š” ์ด์‚ฐ ์ ๋ถ„๊ณ„์˜ ๋Œ€์ˆ˜์ ยท๊ธฐํ•˜ํ•™์  ํŠน์„ฑ์„ ์ดํ•ดํ•˜๊ธฐ ์œ„ํ•ด ๊ฐ€์žฅ ์ผ๋ฐ˜์ ์ธ ๋„ค ํผํ…์…œ Ablowitzโ€‘Ladik (AL) ๊ณ„์ธต์„ ๋Œ€์ƒ์œผ๋กœ ํ•œ๋‹ค. ๊ธฐ์กด ๋ฌธํ—Œ์—์„œ๋Š” ์ฃผ๋กœ ๋‘ ํผํ…์…œ ํ˜•ํƒœ ({Q n,R n})๊ฐ€ ๋‹ค๋ฃจ์–ด์กŒ์œผ๋ฉฐ, ์ด๋Š” ์—ฐ์† AKNSโ€‘ZS ์ŠคํŽ™ํŠธ๋Ÿผ ๋ฌธ์ œ์˜ ์ง์ ‘ ์ด์‚ฐํ™”์— ํ•ด๋‹นํ•œ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜ ๋„ค ํผํ…์…œ ํ™•์žฅ์€ ๋ณด๋‹ค ํ’๋ถ€ํ•œ ์ž์œ ๋„๋ฅผ ์ œ๊ณตํ•˜๊ณ , ๋ณต์žกํ•œ ๋น„์„ ํ˜• ์ฐจ๋ถ„ ๋ฐฉ์ •์‹๋“ค์˜ ๋ผ๊ทธ๋ž‘์ง€์•ˆ ๊ตฌ์กฐ์™€ ๋ณด์กด๋Ÿ‰์„ ํฌ๊ด„์ ์œผ๋กœ ์„ค๋ช…ํ•  ์ˆ˜ ์žˆ๋‹ค. ๋…ผ๋ฌธ์€ ๋จผ์ € ๋“ฑ์ŠคํŽ™ํŠธ๋Ÿผ ํ๋ฆ„๊ณผ ๋น„๋“ฑ์ŠคํŽ™ํŠธ๋Ÿผ ํ๋ฆ„์„ ๋™์ผํ•œ ํ˜•์‹ (u {n,t} L^{m}H^{(0

Nonlinear Sciences
System Virtualization and Efficient ID Transmission Method for RFID Tag   Infrastructure Network

System Virtualization and Efficient ID Transmission Method for RFID Tag Infrastructure Network

: RFID ๊ธฐ์ˆ ์€ ์ •๋ณด ์‚ฌํšŒ์—์„œ ๋ฌผ์ฒด ์‹๋ณ„์— ํ•„์ˆ˜์ ์ธ ์—ญํ• ์„ ํ•˜๋ฉฐ, ์œ ๋น„์ฟผํ„ฐ์Šค ์‚ฌํšŒ๋กœ ๋‚˜์•„๊ฐ์— ๋”ฐ๋ผ ๊ทธ ์ค‘์š”์„ฑ์ด ๋”์šฑ ์ปค์ง€๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์ด์— ๋”ฐ๋ผ RFID ํƒœ๊ทธ ๋„คํŠธ์›Œํฌ ์‹œ์Šคํ…œ์„ ์‚ฌํšŒ์  ์ธํ”„๋ผ๋กœ ๊ตฌ์ถ•ํ•˜๊ธฐ ์œ„ํ•œ ์—ฐ๊ตฌ๊ฐ€ ํ•„์š”ํ•ฉ๋‹ˆ๋‹ค. ๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” RFID ๋„คํŠธ์›Œํฌ์˜ ๊ฐ€์ƒํ™” ๋ฐ ํšจ์œจ์ ์ธ ID ์ „์†ก ๋ฐฉ๋ฒ•์„ ํ†ตํ•ด ์ด๋Ÿฌํ•œ ๋ชฉํ‘œ๋ฅผ ๋‹ฌ์„ฑํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค. RFID ํƒœ๊ทธ ๋„คํŠธ์›Œํฌ ์‹œ์Šคํ…œ์˜ ๊ฐ€์ƒํ™”๋Š” ์—ฌ๋Ÿฌ ์„œ๋น„์Šค ์‹œ์Šคํ…œ์ด ๋™์ผํ•œ ๋ฌผ๋ฆฌ์  ๋„คํŠธ์›Œํฌ๋ฅผ ๊ณต์œ ํ•  ์ˆ˜ ์žˆ๋„๋ก ํ•˜๋Š” ๊ธฐ์ˆ ์ž…๋‹ˆ๋‹ค. ์ด๋Š” ์‹œ์Šคํ…œ ๋น„์šฉ์„ ์ ˆ๊ฐํ•˜๊ณ , ๋ฆฌ๋” ์„ค์น˜ ๊ณต๊ฐ„๊ณผ ์šด์˜ ๋น„์šฉ์„ ํฌ๊ฒŒ ์ค„์ผ

Computer Science Network Networking System
No Image

Taming Serverless Cold Starts Through OS Co-Design

๋ณธ ๋…ผ๋ฌธ์€ ๊ธฐ์กด ์„œ๋ฒ„๋ฆฌ์Šค ์ปดํ“จํŒ… ์‹œ์Šคํ…œ์˜ ์ฝœ๋“œ ์Šคํƒ€ํŠธ ๋ฌธ์ œ๋ฅผ ์‹ฌ์ธต์ ์œผ๋กœ ๋ถ„์„ํ•˜๊ณ , ์ด๋ฅผ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•œ ์ƒˆ๋กœ์šด ์ ‘๊ทผ๋ฒ•์„ ์ œ์‹œํ•œ๋‹ค. OS์™€ ์ง์ ‘ ํ†ตํ•ฉ๋œ Spice๋Š” ์ปค๋„ ์ƒํƒœ ๋ณต์›์„ ์ตœ์ ํ™”ํ•˜์—ฌ ๋น„์šฉ์„ ์ค„์ด๊ณ , ๋ฉ”๋ชจ๋ฆฌ ๋งคํ•‘ ์›์‹œ๊ฐ’์„ ๊ฐœ์„ ํ•จ์œผ๋กœ์จ ์‹ ๋ขฐ์„ฑ๊ณผ ํšจ์œจ์„ฑ์„ ๋†’์˜€๋‹ค. ์ด๋Ÿฌํ•œ ํ˜์‹ ์ ์ธ ์„ค๊ณ„๋Š” ์„œ๋ฒ„๋ฆฌ์Šค ์ปดํ“จํŒ…์˜ ์„ฑ๋Šฅ๊ณผ ํƒ„๋ ฅ์„ฑ ์‚ฌ์ด์˜ ๊ท ํ˜•์„ ๋งž์ถ”๊ณ , ๋” ๋‚˜์€ ์‚ฌ์šฉ์ž ๊ฒฝํ—˜์„ ์ œ๊ณตํ•  ์ˆ˜ ์žˆ๋Š” ์ž ์žฌ๋ ฅ์„ ๋ณด์—ฌ์ค€๋‹ค.

Target tracking in the recommender space: Toward a new recommender   system based on Kalman filtering

Target tracking in the recommender space: Toward a new recommender system based on Kalman filtering

: ์นผ๋งŒ ํ•„ํ„ฐ๋ง ๊ธฐ๋ฐ˜์˜ ์ถ”์ฒœ ์‹œ์Šคํ…œ์€ ๊ธฐ์กด ๋ฐฉ๋ฒ•๋“ค์˜ ํ•œ๊ณ„๋ฅผ ๊ทน๋ณตํ•˜๊ณ , ๊ฐœ์ธํ™” ๋œ ์ถ”์ฒœ์„ ์œ„ํ•œ ์ƒˆ๋กœ์šด ๋Œ€์•ˆ์„ ์ œ์‹œํ•ฉ๋‹ˆ๋‹ค. ๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ์‚ฌ์šฉ์ž๋ฅผ ์ถ”์ ํ•˜๊ณ  ๊ทธ๋“ค์˜ ๋ฏธ๋ž˜ ํ–‰๋™์„ ์˜ˆ์ธกํ•˜๋Š” ๋ฐ ์ง‘์ค‘ํ•˜์—ฌ, ์ถ”์ฒœ ์ •ํ™•๋„๋ฅผ ํ–ฅ์ƒ์‹œํ‚ค๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค. ๋™์  ์ฝ˜ํ…์ธ ๊ฐ€ ํฌํ•จ๋œ ์›น ์„œ๋น„์Šค์—์„œ ๊ฐœ์ธํ™” ๋œ ์ถ”์ฒœ์€ ๊ณ ๊ฐ ๋งŒ์กฑ๊ณผ ์œ ์ง€์— ์ค‘์š”ํ•œ ์—ญํ• ์„ ํ•ฉ๋‹ˆ๋‹ค. ๊ธฐ์กด ํ˜‘์—… ํ•„ํ„ฐ๋ง ๋ฐฉ๋ฒ•์€ ์‚ฌ์šฉ์ž ๊ฐ„์˜ ์œ ์‚ฌ์„ฑ์„ ๊ธฐ๋ฐ˜์œผ๋กœ ํ•˜์ง€๋งŒ, ์ฝœ๋“œ ์Šคํƒ€ํŠธ ๋ฌธ์ œ์™€ ํฌ์†Œ์„ฑ ์ด์Šˆ๋ฅผ ๊ฐ€์ง€๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ๋‹จ์ ์„ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•ด ๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ์นผ๋งŒ ํ•„ํ„ฐ๋ง์„ ๋„์ž…ํ•˜์—ฌ ์‚ฌ์šฉ์ž์˜ ์ƒํƒœ ๊ณต๊ฐ„์„ ๋ชจ๋ธ๋ง

Computer Science System Artificial Intelligence
No Image

Text to model via SysML: Automated generation of dynamical system computational models from unstructured natural language text via enhanced System Modeling Language diagrams

์ด ๋…ผ๋ฌธ์€ ๋™์  ์‹œ์Šคํ…œ ์„ค๊ณ„ ๋ฐ ๋ฐฐํฌ๋ฅผ ๊ฐ€์†ํ™”ํ•˜๊ธฐ ์œ„ํ•œ ํ˜์‹ ์ ์ธ ์ ‘๊ทผ๋ฒ•์„ ์ œ์‹œํ•œ๋‹ค. ์ œ์•ˆ๋œ ์ „๋žต์€ ์ž์—ฐ์–ด ์ฒ˜๋ฆฌ ๊ธฐ์ˆ ๊ณผ ๋Œ€ํ˜• ์–ธ์–ด ๋ชจ๋ธ์˜ ์กฐํ•ฉ์„ ํ™œ์šฉํ•˜์—ฌ ๋ณต์žกํ•œ ๋™์  ์‹œ์Šคํ…œ์˜ ๊ณ„์‚ฐ ๋ชจ๋ธ ์ƒ์„ฑ์„ ์ž๋™ํ™”ํ•˜๋Š” ๋ฐ ์ค‘์ ์„ ๋‘”๋‹ค. SysML ๋‹ค์ด์–ด๊ทธ๋žจ์„ ์‚ฌ์šฉํ•˜์—ฌ ์ •ํ™•ํ•œ ์ •๋ณด๋ฅผ ์ถ”์ถœํ•˜๊ณ , ์ด๋ฅผ ํ†ตํ•ด ์—”์ง€๋‹ˆ์–ด๋ง ํ”„๋กœ์„ธ์Šค๋ฅผ ๊ฐ„์†Œํ™”ํ•˜๊ณ  ๊ฐ€์†ํ™”ํ•  ์ˆ˜ ์žˆ๋‹ค. ๋‹ค์–‘ํ•œ ์‚ฌ๋ก€ ์—ฐ๊ตฌ๋ฅผ ํ†ตํ•ด ์ด ์ ‘๊ทผ๋ฒ•์˜ ์ ์šฉ ๊ฐ€๋Šฅ์„ฑ๊ณผ ์œ ์šฉ์„ฑ์ด ์ž…์ฆ๋˜์—ˆ์œผ๋ฉฐ, ํŠนํžˆ ๋‹จ์ˆœ ํ‰ํ˜•์ถ”์˜ ์˜ˆ์‹œ๋Š” ์ œ์•ˆ๋œ ๋ฐฉ๋ฒ•๋ก ์˜ ํšจ์œจ์„ฑ์„ ๋ณด์—ฌ์ค€๋‹ค. ๋˜ํ•œ, ์ด ๋…ผ๋ฌธ์€ NLP ์ „๋žต๊ณผ LLM์„ ํšจ๊ณผ์ ์œผ๋กœ ํ†ตํ•ฉํ•˜์—ฌ

System Model
The Axial Anomaly and Large Pulsar Kicks

The Axial Anomaly and Large Pulsar Kicks

: ํŽ„์„œ ํ‚ฅ์€ ์ฒœ์ฒด๋ฌผ๋ฆฌํ•™ ๋ถ„์•ผ์—์„œ ์˜ค๋žซ๋™์•ˆ ๋…ผ์˜๋˜์–ด ์˜จ ์ฃผ์ œ์ž…๋‹ˆ๋‹ค. ํŽ„์„œ๋Š” ๋งค์šฐ ๋น ๋ฅธ ์†๋„๋กœ ์ด๋™ํ•˜๋Š” ์ค‘์„ฑ์ž๋ณ„ ๋˜๋Š” ์ฟผํฌ๋ณ„์ด๋ฉฐ, ๊ทธ ๊ธฐ์›๊ณผ ๋ฉ”์ปค๋‹ˆ์ฆ˜์— ๋Œ€ํ•œ ์ดํ•ด๋Š” ์•„์ง ๋ฏธ์™„์„ฑ๋œ ์ƒํƒœ์ž…๋‹ˆ๋‹ค. ์ด ๋…ผ๋ฌธ์—์„œ๋Š” ๊ธฐ์กด ์ด๋ก ๋“ค์ด ์„ค๋ช…ํ•˜์ง€ ๋ชปํ•˜๋Š” ํฐ ํŽ„์„œ ํ‚ฅ์˜ ์›์ธ์„ ์œ„์ƒ ๋ฒกํ„ฐ ์ „๋ฅ˜์—์„œ ์ฐพ์Šต๋‹ˆ๋‹ค. ์œ„์ƒ ๋ฒกํ„ฐ ์ „๋ฅ˜๋Š” ์ค‘์„ฑ์ž๋ณ„๊ณผ ์ฟผํฌ๋ณ„ ๋‚ด๋ถ€์—์„œ ๋ฐœ์ƒํ•  ์ˆ˜ ์žˆ๋Š” ๋…ํŠนํ•œ ํ˜„์ƒ์ž…๋‹ˆ๋‹ค. ์ด๋Š” ์•ฝํ•œ ์ƒํ˜ธ์ž‘์šฉ์— ์˜ํ•ด ๋ฐ˜์ „์„ฑ์ด ์œ„๋ฐ˜๋˜๋Š” ํ™˜๊ฒฝ์—์„œ ์ƒ์„ฑ๋˜๋ฉฐ, ์ฃผ๋กœ ์ขŒํ•ธ ์ž…์ž๋“ค์ด ์šฐ์„ธํ•˜๊ฒŒ ์กด์žฌํ•ฉ๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ์ „๋ฅ˜๋Š” ๋ณ„์˜ ๋‚ด๋ถ€์— ์กด์žฌํ•˜๋ฉฐ, ํŠน์ • ์กฐ๊ฑด์ด ์ถฉ์กฑ๋  ๋•Œ ๊ทธ

Astrophysics HEP-PH
The backbone of the climate network

The backbone of the climate network

: ์ตœ๊ทผ ๋ณต์žกํ•œ ๋„คํŠธ์›Œํฌ ์ด๋ก ์€ ๋‹ค์–‘ํ•œ ๋ถ„์•ผ์—์„œ ๋ณต์žกํ•œ ์‹œ์Šคํ…œ์„ ์ดํ•ดํ•˜๋Š” ๊ฐ•๋ ฅํ•œ ๋„๊ตฌ๋กœ ์ฃผ๋ชฉ๋ฐ›๊ณ  ์žˆ๋‹ค. ํŠนํžˆ ๊ธฐํ›„ ๊ณผํ•™์— ์ ์šฉ๋œ ์—ฐ๊ตฌ๊ฐ€ ์ฆ๊ฐ€ํ•˜๊ณ  ์žˆ๋Š” ๊ฐ€์šด๋ฐ, ๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ๊ธฐ์กด ์„ ํ˜• ๋ฐฉ๋ฒ•์˜ ํ•œ๊ณ„๋ฅผ ๊ทน๋ณตํ•˜๊ธฐ ์œ„ํ•ด ๋น„์„ ํ˜• ์ƒํ˜ธ ์ •๋ณด์™€ ์ค‘์žฌ ์ค‘์‹ฌ์„ฑ(BC)์„ ํ™œ์šฉํ•˜์—ฌ ๊ธฐํ›„ ๋„คํŠธ์›Œํฌ๋ฅผ ๋ถ„์„ํ•œ๋‹ค. ๊ธฐํ›„ ๋„คํŠธ์›Œํฌ๋Š” ๊ณต๊ฐ„ ๊ฒฉ์ž์ ๊ณผ ์ด์ƒ์น˜ ์‹œ๊ฐ„ ์‹œ๋ฆฌ์ฆˆ ์‚ฌ์ด์˜ ํ†ต๊ณ„์  ์ƒํ˜ธ ์˜์กด์„ฑ์„ ๊ธฐ๋ฐ˜์œผ๋กœ ๊ตฌ์ถ•๋œ๋‹ค. ์ด๋Š” ๊ธฐํ›„ ์‹œ์Šคํ…œ์˜ ์—ญํ•™๊ณผ ์œ„์ƒ์„ ์ดํ•ดํ•˜๋Š” ๋ฐ ์ƒˆ๋กœ์šด ํ†ต์ฐฐ๋ ฅ์„ ์ œ๊ณตํ•˜๋ฉฐ, ํŠนํžˆ ์ง€์—ญ ์†์„ฑ๊ณผ ๊ธ€๋กœ๋ฒŒ ์ธก์ •๊ฐ’์— ๋Œ€ํ•œ ์ •๋ณด๋ฅผ ์ค€๋‹ค. ๊ธฐ์กด ์—ฐ๊ตฌ์—์„œ๋Š” ์„ ํ˜•

Physics Network
The communication complexity of non-signaling distributions

The communication complexity of non-signaling distributions

๋ณธ ๋…ผ๋ฌธ์€ ํ†ต์‹  ๋ณต์žก๋„ ์ด๋ก ๊ณผ ์–‘์ž ์ •๋ณด ์ด๋ก  ์‚ฌ์ด์˜ ๊ต์ฐจ์ ์„ ์ฒด๊ณ„์ ์œผ๋กœ ํƒ๊ตฌํ•œ๋‹ค. ๊ธฐ์กด ์—ฐ๊ตฌ๋Š” ์ฃผ๋กœ ํŠน์ • ๋ฌธ์ œ(์˜ˆ: XOR ๊ฒŒ์ž„, ๋ถˆ๋ฆฌ์–ธ ํ•จ์ˆ˜) ํ˜น์€ ํŠน์ • ๋ชจ๋ธ(์˜ˆ: ํ•œ ๋ฒˆ์˜ ๋ผ์šด๋“œ, ๊ณต์œ  ์–ฝํž˜)์—์„œ ๊ฐœ๋ณ„์ ์ธ ํ•˜ํ•œ์„ ์ œ์‹œํ•ด ์™”์œผ๋ฉฐ, ๊ฐ๊ฐ์˜ ๊ธฐ๋ฒ•์ด ์„œ๋กœ ๋‹ค๋ฅธ ์ˆ˜ํ•™์  ๋„๊ตฌ(์˜ˆ: Fourier ๋ถ„์„, ์ •๋ณด ์ด๋ก ์  ๋ถˆ๊ท ํ˜•)์™€ ์—ฐ๊ฒฐ๋ผ ์žˆ์—ˆ๋‹ค. ์ €์ž๋“ค์€ ์ด๋Ÿฌํ•œ ํŒŒํŽธํ™”๋œ ๊ฒฐ๊ณผ๋“ค์„ ํ•˜๋‚˜์˜ ํ†ตํ•ฉ ํ”„๋ ˆ์ž„์›Œํฌ ์•ˆ์œผ๋กœ ๋Œ์–ด๋“ค์ธ๋‹ค. ํ•ต์‹ฌ์€ โ€œaffine ์กฐํ•ฉโ€์ด๋ผ๋Š” ๊ฐœ๋…์ด๋‹ค. ์ €์ž๋“ค์€ ๋ณต์žก๋„๊ฐ€ ๋‚ฎ์€ ๊ธฐ๋ณธ ๋ถ„ํฌ๋“ค์˜ ์„ ํ˜•(์ •ํ™•ํžˆ๋Š” affine) ๊ฒฐํ•ฉ์„ ํ†ตํ•ด

Computer Science Quantum Physics Computational Complexity
No Image

The FM Agent

FM Agent๋Š” LLM์˜ ์ถ”๋ก  ๋Šฅ๋ ฅ๊ณผ ์ง„ํ™”์  ํƒ์ƒ‰์˜ ํšจ์œจ์„ฑ์„ ๊ฒฐํ•ฉํ•˜์—ฌ ๋ณต์žกํ•œ ๋ฌธ์ œ๋ฅผ ํ•ด๊ฒฐํ•˜๋Š” ๊ฐ•๋ ฅํ•œ ๋„๊ตฌ๋ฅผ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค. ํŠนํžˆ, ์ฝœ๋“œ ์Šคํƒ€ํŠธ ์ดˆ๊ธฐํ™” ๋‹จ๊ณ„๋Š” ์ „๋ฌธ๊ฐ€ ์ง€์นจ์„ ํ†ตํ•ด ์—์ด์ „ํŠธ๊ฐ€ ๋น ๋ฅด๊ฒŒ ํ•™์Šตํ•˜๊ณ  ์ ์‘ํ•  ์ˆ˜ ์žˆ๋„๋ก ๋„์™€์ค๋‹ˆ๋‹ค. ๋˜ํ•œ, ์ƒˆ๋กœ์šด ์ง„ํ™”์  ์ƒ˜ํ”Œ๋ง ์ „๋žต์€ ๋ฐ˜๋ณต์ ์ธ ์ตœ์ ํ™”๋ฅผ ๊ฐ€๋Šฅํ•˜๊ฒŒ ํ•˜์—ฌ ๋” ๋‚˜์€ ๊ฒฐ๊ณผ๋ฅผ ์–ป์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ๋„๋ฉ”์ธ๋ณ„ ํ‰๊ฐ€์ž๋Š” ์ •ํ™•์„ฑ๊ณผ ํšจ๊ณผ์„ฑ์„ ๊ณ ๋ คํ•˜์—ฌ ์˜ฌ๋ฐ”๋ฅธ ๋ฐฉํ–ฅ์œผ๋กœ ์—์ด์ „ํŠธ๋ฅผ ์œ ๋„ํ•˜๋ฉฐ, LLM์˜ ํ”ผ๋“œ๋ฐฑ์„ ํ†ตํ•ด ๋”์šฑ ์ •๊ตํ•œ ํƒ์ƒ‰์ด ๊ฐ€๋Šฅํ•ฉ๋‹ˆ๋‹ค. ๋ถ„์‚ฐํ˜• ๋น„๋™๊ธฐ ์‹คํ–‰ ์ธํ”„๋ผ๋Š” ๋ณ‘๋ ฌ ์ฒ˜๋ฆฌ๋ฅผ ํ†ตํ•ด ํšจ์œจ์ ์ธ ๋ฌธ์ œ ํ•ด

No Image

The Formalism-Implementation Gap in Reinforcement Learning Research

๊ฐ•ํ™” ํ•™์Šต ์—ฐ๊ตฌ๋Š” ์ตœ๊ทผ ๋ช‡ ๋…„๊ฐ„ ๋†€๋ผ์šด ๋ฐœ์ „์„ ์ด๋ฃจ์—ˆ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜ ์ด๋Ÿฌํ•œ ๋ฐœ์ „์€ ์ฃผ๋กœ ์—์ด์ „ํŠธ์˜ ์„ฑ๋Šฅ๊ณผ ์ดˆ์ธ์ ์ธ ์ž‘์—… ์ˆ˜ํ–‰ ๋Šฅ๋ ฅ์— ์ง‘์ค‘๋˜์–ด ์™”๋‹ค. ์ด ๋…ผ๋ฌธ์—์„œ๋Š” ์ด๋Ÿฌํ•œ ์ถ”์„ธ๊ฐ€ ๊ฐ•ํ™” ํ•™์Šต์˜ ๊ทผ๋ณธ์ ์ธ ์ดํ•ด์™€ ๊ณผํ•™์  ๋ฐœ์ „์— ๋ฏธ์น˜๋Š” ์˜ํ–ฅ์„ ๋ถ„์„ํ•œ๋‹ค. RL ์—ฐ๊ตฌ๊ฐ€ ๋ฒค์น˜๋งˆํฌ ๊ณผ์ œ์— ๋„ˆ๋ฌด ์ง‘์ค‘ํ•˜๊ฒŒ ๋˜๋ฉด, ์‹ค์ œ ์„ธ๊ณ„ ๋ฌธ์ œ์— ๋Œ€ํ•œ ์ ์šฉ ๊ฐ€๋Šฅ์„ฑ์„ ์ œํ•œํ•  ์ˆ˜ ์žˆ๋‹ค. ๋˜ํ•œ, ์„ฑ๋Šฅ์„ ํ–ฅ์ƒํ•˜๋Š” ๋ฐ์—๋งŒ ์ดˆ์ ์„ ๋งž์ถ”๋ฉด RL ๊ธฐ๋ฒ•์˜ ๋‚ด์žฌ๋œ ํ•™์Šต ์—ญํ•™์— ๋Œ€ํ•œ ์ดํ•ด๋ฅผ ๋†“์น  ์ˆ˜ ์žˆ๋‹ค. ์ด ๋…ผ๋ฌธ์€ ๊ฐ•ํ™” ํ•™์Šต ์—ฐ๊ตฌ์˜ ๋ฐฉํ–ฅ์„ ์ „ํ™˜ํ•˜์—ฌ ๋” ๊นŠ์€ ์ดํ•ด์™€ ๊ณผํ•™์  ๋ฐœ์ „์œผ๋กœ ๋‚˜์•„๊ฐ€์•ผ

Learning
The Fundamentals of Policy Crowdsourcing

The Fundamentals of Policy Crowdsourcing

(800์ž ์ด์ƒ) ๋ณธ ๋…ผ๋ฌธ์€ ์ •์ฑ… ๋ถ„์•ผ์— ํฌ๋ผ์šฐ๋“œ์†Œ์‹ฑ์„ ์ฒด๊ณ„์ ์œผ๋กœ ์ ์šฉํ•˜๊ธฐ ์œ„ํ•œ โ€˜ํ†ตํ•ฉ ํ”„๋ ˆ์ž„์›Œํฌโ€™๋ฅผ ์ œ์‹œํ•œ๋‹ค๋Š” ์ ์—์„œ ํ•™์ˆ ์ ยท์‹ค๋ฌด์  ์˜์˜๊ฐ€ ํฌ๋‹ค. ๋จผ์ € ์ €์ž๋Š” ๊ธฐ์กด ํฌ๋ผ์šฐ๋“œ์†Œ์‹ฑ ์—ฐ๊ตฌ์—์„œ ๋„์ถœ๋œ ๋‘ ๊ฐ€์ง€ ์œ ํ˜•๋ก โ€”โ€˜์ž‘์—… ๊ธฐ๋ฐ˜ ๋ชจ๋ธ(Production Model)โ€™๊ณผ โ€˜์•„์ด๋””์–ด ๊ณต๋ชจ ๋ชจ๋ธ(Idea Competition Model)โ€™โ€”์„ ์ •์ฑ… ์‚ฌ์ดํด์— ๋งคํ•‘ํ•œ๋‹ค. ์ด๋ฅผ ์œ„ํ•ด ๊ฐ€์ƒ๋…ธ๋™์‹œ์žฅ(VLM), ํ† ๋„ˆ๋จผํŠธ ํฌ๋ผ์šฐ๋“œ์†Œ์‹ฑ(TC), ๊ฐœ๋ฐฉํ˜• ํ˜‘์—…(Open Collaboration)์ด๋ผ๋Š” ์„ธ ๊ฐ€์ง€ ๋Œ€ํ‘œ์  ๋ฉ”์ปค๋‹ˆ์ฆ˜์„ ์„ ์ •ํ•˜๊ณ , ๊ฐ๊ฐ์„ 7๊ฐ€์ง€ ํŠน์„ฑ(๋ณด

Computer Science Social Networks Computers and Society HCI
No Image

The Geometric View of Theories

์ด ๋…ผ๋ฌธ์€ ๊ณผํ•™ ์ด๋ก ์„ ์ดํ•ดํ•˜๋Š” ์ƒˆ๋กœ์šด ์ ‘๊ทผ๋ฒ•์„ ์ œ์‹œํ•˜๋ฉฐ, ํŠนํžˆ ๋ฌผ๋ฆฌํ•™์—์„œ ๋ฐœ๊ฒฌ๋˜๋Š” ์ด์ค‘์„ฑ๊ณผ ์ค€ ์ด์ค‘์„ฑ์— ์ดˆ์ ์„ ๋งž์ถ˜๋‹ค. ์ €์ž๋Š” ์ด๋ก ์„ ๋‹จ์ˆœํ•œ ๋ชจ๋ธ๋“ค์˜ ์ง‘ํ•ฉ์œผ๋กœ ๋ณด๋Š” ๊ฒƒ์„ ๋„˜์–ด์„œ์„œ ๊ธฐํ•˜ํ•™์  ๊ด€์ ์—์„œ ๋ฐ”๋ผ๋ณธ๋‹ค. ์ด๋Š” ์ด๋ก ์˜ ๊ตฌ์กฐ์™€ ๊ทธ ๋‚ด์žฌ๋œ ์œ„์ƒ์ˆ˜ํ•™์  ๋ฐ ๊ธฐํ•˜ํ•™์  ํŠน์„ฑ์„ ๊ฐ•์กฐํ•œ๋‹ค. ์ œ์•ˆ๋œ '๋ชจ๋ธ ๋ฒˆ๋“ค' ๊ฐœ๋…์€ ๋‹ค์–‘ํ•œ ์ด๋ก ๋“ค์„ ํ•˜๋‚˜์˜ ๊ตฌ์กฐ๋กœ ํ†ตํ•ฉํ•˜๋ฉฐ, ์ค€ ์ด์ค‘์„ฑ์ด ์„ฌ์œ  ์‚ฌ์ด์˜ ์ „์ด ํ•จ์ˆ˜๋กœ์„œ ์ค‘์š”ํ•œ ์—ญํ• ์„ ํ•œ๋‹ค๋Š” ๊ฒƒ์„ ๋ณด์—ฌ์ค€๋‹ค. ๋˜ํ•œ, ์ด ๊ด€์ ์€ ๋ฌผ๋ฆฌ ์ด๋ก ์— ๋Œ€ํ•œ ํ˜„์‹ค์ฃผ์˜ ํ•ด์„์„ ํฌํ•จํ•œ ์—ฌ๋Ÿฌ ์ฒ ํ•™์  ๋ฌธ์ œ๋“ค์„ ์ œ๊ธฐํ•œ๋‹ค.

The Geometry of Culture: Analyzing Meaning through Word Embeddings

The Geometry of Culture: Analyzing Meaning through Word Embeddings

๋ณธ ๋…ผ๋ฌธ์€ ๋ฌธํ™”์‚ฌํšŒํ•™์—์„œ ์ „ํ†ต์ ์œผ๋กœ ์‚ฌ์šฉ๋˜์–ด ์˜จ ํ…์ŠคํŠธ ๋ถ„์„ ๋ฐฉ๋ฒ•์˜ ํ•œ๊ณ„๋ฅผ ์‹ ๊ฒฝ๋ง ๊ธฐ๋ฐ˜ ๋‹จ์–ด ์ž„๋ฒ ๋”ฉ์ด๋ผ๋Š” ์ตœ์‹  ์ž์—ฐ์–ด ์ฒ˜๋ฆฌ(NLP) ๊ธฐ์ˆ ๋กœ ๊ทน๋ณตํ•˜๊ณ ์ž ํ•œ๋‹ค. ๋จผ์ € ์ €์ž๋Š” ๋””์ง€ํ„ธํ™”๋œ ๋ฐฉ๋Œ€ํ•œ ํ…์ŠคํŠธ๊ฐ€ ์‚ฌํšŒ์  ํ–‰์œ„ยท์ธ์ง€ยท๊ฐ์ • ๋“ฑ์„ ํฌ๊ด„ํ•œ๋‹ค๋Š” ์ ์„ ๊ฐ•์กฐํ•˜๊ณ , ์ด๋Ÿฌํ•œ ๋ฐ์ดํ„ฐ๊ฐ€ ๊ธฐ์กด์˜ LSAยทLSI์™€ ๊ฐ™์€ ํ†ต๊ณ„์  ๋ฐฉ๋ฒ•์œผ๋กœ๋Š” ์ถฉ๋ถ„ํžˆ ํ™œ์šฉ๋˜์ง€ ๋ชปํ•œ๋‹ค๋Š” ๋น„ํŒ์„ ์ œ์‹œํ•œ๋‹ค. ์ž„๋ฒ ๋”ฉ ๋ชจ๋ธ์€ ๋‹จ์–ด๋ฅผ ๊ณ ์ฐจ์› ์‹ค์ˆ˜ ๋ฒกํ„ฐ๋กœ ๋งคํ•‘ํ•˜๊ณ , ์ฃผ๋ณ€ ๋‹จ์–ด(์ปจํ…์ŠคํŠธ) ์ •๋ณด๋ฅผ ํ•™์Šตํ•จ์œผ๋กœ์จ ์˜๋ฏธ์  ์œ ์‚ฌ์„ฑ์„ ๊ณต๊ฐ„์  ๊ฑฐ๋ฆฌ๋กœ ํ‘œํ˜„ํ•œ๋‹ค. ํŠนํžˆ โ€œ๋‚จ์„ฑโ€‘์—ฌ์„ฑโ€, โ€œ๋ถ€์œ โ€‘๋นˆ๊ณคโ€ ๋“ฑ ์‚ฌํšŒยท

Computer Science NLP
The Hidden Cost of Straight Lines: Quantifying Misallocation Risk in Voronoi-based Service Area Models

The Hidden Cost of Straight Lines: Quantifying Misallocation Risk in Voronoi-based Service Area Models

๋ณธ ๋…ผ๋ฌธ์€ ์œ ํด๋ฆฌ๋“œ ๊ฑฐ๋ฆฌ๊ฐ€ ๊ธฐ๋Šฅ์  ์ ‘๊ทผ์„ฑ์„ ์ œ๋Œ€๋กœ ๋Œ€๋ณ€ํ•˜์ง€ ๋ชปํ•œ๋‹ค๋Š” ๊ฐ€์ •์—์„œ ์ถœ๋ฐœํ•˜์—ฌ, ๋ณด๋กœ๋…ธ์ด ๊ธฐ๋ฐ˜ ์„œ๋น„์Šค ์˜์—ญ ๋ชจ๋ธ์˜ ํ•œ๊ณ„๋ฅผ ์‹ค์ฆ์ ์œผ๋กœ ๋ณด์—ฌ์ฃผ๊ณ  ์ด๋ฅผ ์ •๋Ÿ‰ํ™”ํ•˜๋Š” ํ™•๋ฅ ๋ก ์  ํ‹€์„ ๊ฐœ๋ฐœํ•ฉ๋‹ˆ๋‹ค. ์ด ํ‹€์€ ์œ ํด๋ฆฌ๋“œ ๊ฑฐ๋ฆฌ์™€์˜ ๋ถˆ์ผ์น˜๋ฅผ ๋กœ๊ทธ์ •๊ทœ ๋ถ„ํฌ๋ฅผ ํ†ตํ•ด ๋ชจ๋ธ๋งํ•˜๊ณ , ์ง€์—ญ ๊ธฐํ•˜ํ•™์— ๋”ฐ๋ผ ํ• ๋‹น ์˜ค๋ฅ˜ ํ™•๋ฅ ์„ ์œ ๋„ํ•ฉ๋‹ˆ๋‹ค. ์ŠคํŽ˜์ธ ์—‘์ŠคํŠธ๋ ˆ๋งˆ๋‘๋ผ ์ง€์—ญ์˜ ๊ณต์žฅ ์ง€์ž์ฒด ๊ด€์ธก์น˜(n 383)๋ฅผ ์‚ฌ์šฉํ•˜์—ฌ, ์ด ํ‹€์ด 15.4%์˜ ์ง€์ž์ฒด๊ฐ€ ๊ธฐ๋Šฅ์ ์œผ๋กœ ์˜คํ• ๋‹น๋˜์—ˆ์Œ์„ ๋ณด์—ฌ์ค๋‹ˆ๋‹ค. ์ด๋Š” ์ด๋ก ์  ์˜ˆ์ธก ๊ตฌ๊ฐ„๊ณผ ์ผ์น˜ํ•˜๋ฉฐ, ๋ณต์žกํ•œ ๊ณต๊ฐ„ ๋ชจ๋ธ๊ณผ 95%์˜ ์ผ์น˜๋ฅผ ๋‹ฌ์„ฑํ•ฉ๋‹ˆ๋‹ค.

Model
The Information Content of Sarbanes-Oxley in Predicting Security   Breaches

The Information Content of Sarbanes-Oxley in Predicting Security Breaches

์ด ๋…ผ๋ฌธ์€ ํšŒ๊ณ„ยท๊ฐ์‚ฌ ๋ถ„์•ผ์—์„œ ํ”ํžˆ ๋‹ค๋ฃจ๋Š” SOX 302ยท404 ๋ณด๊ณ ์„œ๊ฐ€ ์‹ค์ œ ์‚ฌ์ด๋ฒ„ ๋ณด์•ˆ ์œ„ํ—˜์„ ์˜ˆ์ธกํ•˜๋Š” ๋ฐ ์–ผ๋งˆ๋‚˜ ์œ ์šฉํ•œ์ง€๋ฅผ ์‹ค์ฆ์ ์œผ๋กœ ๊ฒ€์ฆํ•œ ์ ์—์„œ ํ•™์ˆ ์ ยท์‹ค๋ฌด์  ์˜์˜๊ฐ€ ํฌ๋‹ค. ์ฒซ์งธ, ๋ฐ์ดํ„ฐ ์ˆ˜์ง‘ ๋ฐฉ์‹์ด โ€˜๊ณต๊ฐœ๋œ ๋ณด์•ˆ ์นจํ•ด ์‚ฌ๋ก€โ€™๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ํ–ˆ๋‹ค๋Š” ์ ์€ ๋ณด๊ณ  ํŽธํ–ฅ์„ ์ตœ์†Œํ™”ํ•˜๋ ค๋Š” ์‹œ๋„๋กœ ๋ณผ ์ˆ˜ ์žˆ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜ ์นจํ•ด ์‚ฌ๊ฑด ์ž์ฒด๊ฐ€ ๊ธฐ์—…์— ์˜ํ•ด ์„ ํƒ์ ์œผ๋กœ ๊ณต๊ฐœ๋˜๋Š” ํŠน์„ฑ์ƒ, ์—ฌ์ „ํžˆ ํ‘œ๋ณธ์ด ์ œํ•œ์ ์ผ ๊ฐ€๋Šฅ์„ฑ์ด ์žˆ๋‹ค. ๋‘˜์งธ, ์—ฐ๊ตฌ๋Š” ๋‘ ๊ฐ€์ง€ ํ†ต๊ณ„ ๋ชจ๋ธ์„ ๋„์ž…ํ•œ๋‹ค. ์œ„ํ—˜(hazard) ๋ชจ๋ธ์€ ์‹œ๊ฐ„โ€‘์œ„ํ—˜ ๊ด€์ ์—์„œ ํ†ต์ œ ํšจ์œจ์„ฑ์ด ์นจํ•ด ๋ฐœ์ƒ์— ๋ฏธ์น˜๋Š”

Quantitative Finance
No Image

The jet schemes of the nilpotent cone of $mathfrak{gl}_n$ over $mathbb{F}_ell$ and analytic properties of the Chevalley map

1. ์—ฐ๊ตฌ ๋ฐฐ๊ฒฝ ๋ฐ ๋™๊ธฐ Chevalley ์‚ฌ์ƒ ์€ ๊ณ ์ „์ ์ธ ๋ฆฌ๊ตฐ ์ด๋ก ์—์„œ ํ–‰๋ ฌ์˜ ๊ณ ์œ ๋‹คํ•ญ์‹(ํŠน์„ฑ๋‹คํ•ญ์‹)๊ณผ ์—ฐ๊ฒฐ๋˜๋Š” ๊ธฐ๋ณธ ์‚ฌ์ƒ์ด๋ฉฐ, ๊ทธ ํ‘ธ์‹œํฌ์›Œ๋“œ ์ธก์ •์˜ ์ •๊ทœ์„ฑ์€ Harishโ€‘Chandra ์ ๋ถ„์ •๋ฆฌ ์™€ ๊ถค๋„ ์ ๋ถ„ ์˜ ํ•ต์‹ฌ์ด๋‹ค. 0ํŠน์„ฑ์—์„œ๋Š” Springer ํ•ด์ƒ, Grauertโ€‘Riemenschneider ์ •๋ฆฌ, ๊ทธ๋ฆฌ๊ณ  Jordanโ€‘Chevalley ๋ถ„ํ•ด ๋“ฑ์„ ํ™œ์šฉํ•ด Chevalley ์‚ฌ์ƒ์ด FRS (Flat, Reduced, Rationalโ€‘Singular)์ž„์„ ๋ณด์ด๊ณ , ๊ฒฐ๊ณผ์ ์œผ๋กœ (p (mu))๊ฐ€ ์—ฐ์†ํ•จ์ˆ˜์™€ Haar ์ธก์ •์˜

Mathematics
The Malicious Use of Artificial Intelligence: Forecasting, Prevention,   and Mitigation

The Malicious Use of Artificial Intelligence: Forecasting, Prevention, and Mitigation

๋ณธ ๋ณด๊ณ ์„œ๋Š” ๊ธ‰์†ํžˆ ๋ฐœ์ „ํ•˜๋Š” ์ธ๊ณต์ง€๋Šฅ ๊ธฐ์ˆ ์ด ๊ธฐ์กด ๋ณด์•ˆ ํŒจ๋Ÿฌ๋‹ค์ž„์„ ์–ด๋–ป๊ฒŒ ๋’คํ”๋“ค ์ˆ˜ ์žˆ๋Š”์ง€๋ฅผ ์ฒด๊ณ„์ ์œผ๋กœ ์ •๋ฆฌํ•œ๋‹ค๋Š” ์ ์—์„œ ํ•™์ˆ ยท์ •์ฑ… ์–‘์ธก ๋ชจ๋‘์—๊ฒŒ ์ค‘์š”ํ•œ ์ฐธ๊ณ ๋ฌธํ—Œ์ด ๋œ๋‹ค. ์ฒซ์งธ, ์œ„ํ˜‘ ์ง€ํ˜•์˜ 3๊ฐ€์ง€ ๋ณ€ํ™”๋ฅผ ๋ช…ํ™•ํžˆ ๊ตฌ๋ถ„ ํ•œ๋‹ค๋Š” ์ ์ด ๋‹๋ณด์ธ๋‹ค. ๊ธฐ์กด ์œ„ํ˜‘์˜ ํ™•๋Œ€๋Š” AI๊ฐ€ ์ธ๊ฐ„์˜ ๋…ธ๋™ยท์ง€์‹ยท์ „๋ฌธ์„ฑ์„ ๋Œ€์ฒดํ•จ์œผ๋กœ์จ ๊ณต๊ฒฉ ๋น„์šฉ์„ ํ˜„์ €ํžˆ ๋‚ฎ์ถ”๊ณ , ๋”ฐ๋ผ์„œ ๊ณต๊ฒฉ์ž ์ˆ˜์™€ ๊ณต๊ฒฉ ์†๋„๊ฐ€ ๊ธ‰์ฆํ•œ๋‹ค๋Š” ๋…ผ๋ฆฌ๋‹ค. ์ด๋Š” โ€˜AIโ€‘์ž๋™ํ™” โ†’ ๋น„์šฉ ์ ˆ๊ฐ โ†’ ๊ณต๊ฒฉ์ž ํ™•๋Œ€โ€™๋ผ๋Š” ์ธ๊ณผ๊ด€๊ณ„๋ฅผ ๊ตฌ์ฒด์ ์ธ ์‚ฌ๋ก€(์Šคํ”ผ์–ด ํ”ผ์‹ฑ ์ž๋™ํ™”, ์•…์„ฑ ์ฝ”๋“œ ๋ณ€ํ˜• ์ž๋™ ์ƒ์„ฑ ๋“ฑ)์™€ ํ•จ๊ป˜ ์ œ์‹œํ•จ์œผ

Computer Science Artificial Intelligence Computers and Society Cryptography and Security
The Measurement of the Hubble Constant H_0 in the Solar System

The Measurement of the Hubble Constant H_0 in the Solar System

์ด ์—ฐ๊ตฌ๋Š” ์ „ํ†ต์ ์ธ ์ฒœ๋ฌธํ•™์  ๊ฑฐ๋ฆฌ ์‚ฌ๋‹ค๋ฆฌ์™€๋Š” ์ „ํ˜€ ๋‹ค๋ฅธ ์ ‘๊ทผ๋ฒ•์„ ์ œ์‹œํ•œ๋‹ค๋Š” ์ ์—์„œ ํ˜์‹ ์ ์ด๋‹ค. ํ—ˆ๋ธ” ์ƒ์ˆ˜๋Š” ๋ณดํ†ต ์ˆ˜์‹ญ ๋ฉ”๊ฐ€ํŒŒ์„น ์ด์ƒ ๊ฑฐ๋ฆฌ์˜ ์€ํ•˜๋‚˜ ์ดˆ์‹ ์„ฑ, ํ˜น์€ ์ค‘๋ ฅํŒŒ ํ‘œ์ค€์‹ ํ˜ธ๋ฅผ ์ด์šฉํ•ด ์ถ”์ •ํ•œ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜ ์ €์ž๋“ค์€ โ€œ๊ทน๋‹จ์ ์œผ๋กœ ์งง์€ ๊ฑฐ๋ฆฌโ€์ธ ํƒœ์–‘๊ณ„ ๊ทœ๋ชจ(์ˆ˜์ฒœ AU ์ดํ•˜)์—์„œ ๋น›์˜ ํŒŒ์žฅ์ด ์šฐ์ฃผ ํŒฝ์ฐฝ์— ์˜ํ•ด ๋ฏธ์„ธํ•˜๊ฒŒ ๋Š˜์–ด๋‚˜๋Š” ํ˜„์ƒ์„ ์ง์ ‘ ์ธก์ •ํ•œ๋‹ค๋Š” ์ „์ œ์— ๊ธฐ๋ฐ˜ํ•œ๋‹ค. ํ•ต์‹ฌ ์•„์ด๋””์–ด๋Š” ๋‹ค์Œ๊ณผ ๊ฐ™๋‹ค. FLRW ๋ฉ”ํŠธ๋ฆญ์— ์˜ํ•ด ์ •์˜๋˜๋Š” ์Šค์ผ€์ผ ํŒฉํ„ฐ (a(t))๊ฐ€ ์‹œ๊ฐ„์— ๋”ฐ๋ผ ๋ณ€ํ•˜๋ฉด, ์ „์ž๊ธฐ ํŒŒ๋™์˜ ์ฃผํŒŒ์ˆ˜๋Š” ( Delta f/f t,H {

Physics System Astrophysics
The multiple viewpoints as approach to information retrieval within   collaborative development context

The multiple viewpoints as approach to information retrieval within collaborative development context

๋ณธ ์—ฐ๊ตฌ๋Š” ํ˜‘์—… ์ œํ’ˆ ๊ฐœ๋ฐœ์—์„œ ๋ฐœ์ƒํ•˜๋Š” ์ •๋ณด ๊ณผ๋ถ€ํ•˜์™€ ์ดํ•ด๊ด€๊ณ„์ž ๊ฐ„ ์˜๋ฏธ ๋ถˆ์ผ์น˜๋ฅผ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•œ ๊ตฌ์กฐ์  ์ ‘๊ทผ๋ฒ•์„ ์ œ์‹œํ•œ๋‹ค๋Š” ์ ์—์„œ ์˜์˜๊ฐ€ ํฌ๋‹ค. ์ฒซ์งธ, ์ €์ž๋“ค์€ ๊ด€์ (viewpoint) ์„ โ€œ์ •๋ณด ๊ฐ์ฒด ์ง‘ํ•ฉ์„ ํ‘œํ˜„ํ•˜๋Š” ์ฒด๊ณ„์™€ ํ•ด๋‹น ์ •๋ณด๋ฅผ ์ ‘๊ทผํ•˜๊ธฐ ์œ„ํ•œ ๋ฉ”์ปค๋‹ˆ์ฆ˜โ€์œผ๋กœ ์ •์˜ํ•˜๊ณ , ์ด๋ฅผ ๊ธฐ์กด HCIยท์‹œ๊ฐํ™” ๋ถ„์•ผ์—์„œ ์‚ฌ์šฉ๋˜๋Š” ์ •์˜์™€ ๋น„๊ตํ•œ๋‹ค. ์ด ์ •์˜๋Š” ๋‹จ์ˆœํžˆ ์‹œ๊ฐ์  ํ‘œํ˜„์„ ๋„˜์–ด, ์ •๋ณด์˜ ์˜๋ฏธ์  ๊ตฌ์กฐ์™€ ์ ‘๊ทผ ๊ถŒํ•œ๊นŒ์ง€ ํฌ๊ด„ํ•œ๋‹ค๋Š” ์ ์—์„œ ์ฐจ๋ณ„ํ™”๋œ๋‹ค. ๋‘˜์งธ, ๋‹ค์ค‘ ๊ด€์  ์‹œ์Šคํ…œ์ด ์ œ์‹œํ•˜๋Š” ํ•ต์‹ฌ ๋ฉ”์ปค๋‹ˆ์ฆ˜์€ ๊ด€์‹ฌยท๋ชฉํ‘œ ๊ธฐ๋ฐ˜ ํ•„ํ„ฐ๋ง ์ด๋‹ค. ์ดํ•ด๊ด€๊ณ„์ž๋Š”

Computer Science HCI
The origin and propagation of variability in the outflows of long   duration gamma-ray bursts

The origin and propagation of variability in the outflows of long duration gamma-ray bursts

: ๊ฐ๋งˆ์„  ํญ๋ฐœ(GRBs)์€ ์šฐ์ฃผ์—์„œ ๊ฐ€์žฅ ๊ฐ•๋ ฅํ•œ ์‚ฌ๊ฑด ์ค‘ ํ•˜๋‚˜์ด๋ฉฐ, ๊ทธ ๋ณ€๋™์„ฑ๊ณผ ๋ฐฉ์ถœ ํŠน์„ฑ์€ ๊ณผํ•™์ž๋“ค์—๊ฒŒ ๋งŽ์€ ๊ด€์‹ฌ์„ ๋Œ์–ด์™”์Šต๋‹ˆ๋‹ค. ์ด ๋…ผ๋ฌธ์—์„œ๋Š” GRB์˜ ๋ณ€๋™์„ฑ์„ ์‹ฌ๋„ ์žˆ๊ฒŒ ๋ถ„์„ํ•˜๊ณ , ๊ทธ ์›์ธ๊ณผ ์ „ํŒŒ ๊ณผ์ •์„ ์ดํ•ดํ•˜๊ธฐ ์œ„ํ•œ ์—ฐ๊ตฌ๋ฅผ ์ง„ํ–‰ํ•ฉ๋‹ˆ๋‹ค. ๋จผ์ €, ์—ฐ๊ตฌ์ง„์€ GRBs๊ฐ€ ๋‘ ๊ฐ€์ง€ ์œ ํ˜•์œผ๋กœ ๋ถ„๋ฅ˜๋œ๋‹ค๋Š” ๊ฒƒ์„ ์„ค๋ช…ํ•ฉ๋‹ˆ๋‹ค: ์งง์€ ์ง€์† ์‹œ๊ฐ„๊ณผ ๊ธด ์ง€์† ์‹œ๊ฐ„์ž…๋‹ˆ๋‹ค. ๊ธด ์ง€์† ์‹œ๊ฐ„์˜ GRBs๋Š” ๋Œ€์งˆ๋Ÿ‰ ๋ณ„์˜ ์ฃฝ์Œ๊ณผ ๊ด€๋ จ์ด ์žˆ์œผ๋ฉฐ, ์ด๋Ÿฌํ•œ ๋ฐœ๊ฒฌ์€ ํญ๋ฐœ์˜ ๋ณ€๋™์„ฑ์— ๋Œ€ํ•œ ์ƒˆ๋กœ์šด ๊ด€์ ์„ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค. GRB์˜ ๋ณ€๋™์„ฑ์€ ์—”์ง„ ์ž์ฒด์™€ ์ œํŠธ๊ฐ€ ์ „์กฐ ๋ณ„ ๋ฌผ์งˆ๊ณผ

Astrophysics
The price of ignorance: The impact of side-information on delay for   lossless source-coding

The price of ignorance: The impact of side-information on delay for lossless source-coding

์ด ์—ฐ๊ตฌ๋Š” ์ „ํ†ต์ ์ธ ๋ธ”๋กโ€‘์ฝ”๋”ฉ ํŒจ๋Ÿฌ๋‹ค์ž„์„ ๋„˜์–ด, ์‹ค์‹œ๊ฐ„ ์ŠคํŠธ๋ฆผ ํ™˜๊ฒฝ์—์„œ โ€œ์ง€์—ฐโ€์ด๋ผ๋Š” ์ƒˆ๋กœ์šด ์„ฑ๋Šฅ ์ง€ํ‘œ๋ฅผ ๋„์ž…ํ•จ์œผ๋กœ์จ ์ •๋ณด ์ด๋ก ์— ์ƒˆ๋กœ์šด ์‹œ๊ฐ์„ ์ œ๊ณตํ•œ๋‹ค. ๊ธฐ์กด ๋ฌธํ—Œ(Shannon, Slepianโ€‘Wolf ๋“ฑ)์€ ์ธ์ฝ”๋”๊ฐ€ ๋ถ€์ˆ˜ ์ •๋ณด๋ฅผ ๊ฐ–๊ฑฐ๋‚˜ ํ”ผ๋“œ๋ฐฑ์„ ์‚ฌ์šฉํ•˜๋”๋ผ๋„ ์ „์†ก๋ฅ  ํ•œ๊ณ„(์ฑ„๋„ ์šฉ๋Ÿ‰ยท์กฐ๊ฑด๋ถ€ ์—”ํŠธ๋กœํ”ผ)์—๋Š” ์˜ํ–ฅ์„ ๋ฏธ์น˜์ง€ ์•Š๋Š”๋‹ค๊ณ  ์ฃผ์žฅํ•œ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜ ์ด๋Ÿฌํ•œ ๊ฒฐ๊ณผ๋Š” ๋Œ€๋ถ€๋ถ„ ๋ธ”๋ก ๊ธธ์ด๋ฅผ ์ง€์—ฐ์˜ ๋Œ€๋ฆฌ ๋ณ€์ˆ˜๋กœ ์‚ผ์•„, ์ „์ฒด ๋ฉ”์‹œ์ง€๊ฐ€ ํ•œ ๋ฒˆ์— ์ œ๊ณต๋˜๋Š” ์ƒํ™ฉ์„ ์ „์ œ๋กœ ํ•œ๋‹ค. ๋ณธ ๋…ผ๋ฌธ์€ ์†Œ์Šค ์‹ฌ๋ณผ์ด ์‹œ๊ฐ„์— ๋”ฐ๋ผ ์ˆœ์ฐจ์ ์œผ๋กœ ๋ฐœ์ƒํ•˜๊ณ , ๋””์ฝ”๋”๊ฐ€ ์ผ์ •ํ•œ

Computer Science Information Theory Mathematics
The Sample Complexity of Dictionary Learning

The Sample Complexity of Dictionary Learning

๋ณธ ๋…ผ๋ฌธ์€ ๋”•์…”๋„ˆ๋ฆฌ ํ•™์Šต(dictionary learning) ๋ฌธ์ œ๋ฅผ ์ƒ˜ํ”Œ ๋ณต์žก๋„(sample complexity) ๊ด€์ ์—์„œ ์ฒด๊ณ„์ ์œผ๋กœ ๋ถ„์„ํ•œ๋‹ค๋Š” ์ ์—์„œ ์˜๋ฏธ๊ฐ€ ํฌ๋‹ค. ๊ธฐ์กด ์—ฐ๊ตฌ๋“ค์€ ์ฃผ๋กœ ์•Œ๊ณ ๋ฆฌ์ฆ˜์  ์ˆ˜๋ ด ์†๋„๋‚˜ ์žฌ๊ตฌ์„ฑ ์˜ค์ฐจ์— ์ดˆ์ ์„ ๋งž์ถ”์—ˆ์ง€๋งŒ, ์—ฌ๊ธฐ์„œ๋Š” ์ผ๋ฐ˜ํ™”(generalization) โ€”์ฆ‰, ํ•™์Šต์— ์‚ฌ์šฉ๋œ ์œ ํ•œ ์ƒ˜ํ”Œ ์ง‘ํ•ฉ์œผ๋กœ๋ถ€ํ„ฐ ์–ป์€ ์‚ฌ์ „์ด ์ƒˆ๋กœ์šด, ๋ณด์ง€ ๋ชปํ•œ ์‹ ํ˜ธ์— ๋Œ€ํ•ด ์–ด๋А ์ •๋„์˜ ์˜ค๋ฅ˜๋ฅผ ๋ณด์ผ์ง€๋ฅผ ์ด๋ก ์ ์œผ๋กœ ์ •๋Ÿ‰ํ™”ํ•œ๋‹ค. 1. ๋ฌธ์ œ ์„ค์ •๊ณผ ํ•ต์‹ฌ ๊ฐ€์ • ์‹ ํ˜ธ ๊ณต๊ฐ„ (X mathbb{R}^{n}) ์—์„œ ์‹ ํ˜ธ (x)๋Š” ์‚ฌ์ „

Computer Science Statistics Learning Machine Learning
Theory and design of a phase-inverted balanced coupled-line DC-blocker

Theory and design of a phase-inverted balanced coupled-line DC-blocker

๋ณธ ์—ฐ๊ตฌ๋Š” ๊ณ ์ฃผํŒŒ ํšŒ๋กœ ์„ค๊ณ„์—์„œ ์˜ค๋ž˜๋œ ๋ฌธ์ œ์ธ DCโ€‘๋ธ”๋กœํ‚น์„ ์ƒˆ๋กœ์šด ๊ด€์ ์—์„œ ์ ‘๊ทผํ•œ๋‹ค๋Š” ์ ์—์„œ ํ•™์ˆ ์ ยท์‹ค์šฉ์  ์˜์˜๊ฐ€ ํฌ๋‹ค. ๊ธฐ์กด ๊ฒฐํ•ฉ์„  DCโ€‘๋ธ”๋กœ์ปค๋Š” ์งยทํ™€ ๋ชจ๋“œ ์ž„ํ”ผ๋˜์Šค๋ฅผ ์ด์šฉํ•ด ์„ค๊ณ„๋˜์—ˆ์œผ๋ฉฐ, ์„ค๊ณ„ ๊ณผ์ •์ด ๋ณต์žกํ•˜๊ณ  ์ œ์กฐ ๊ณต์ฐจ์— ๋ฏผ๊ฐํ•œ ๋‹จ์ ์ด ์žˆ์—ˆ๋‹ค. ์ €์ž๋“ค์€ ์ „์†ก์„ ์˜ ๊ธฐ๋ณธ ํŒŒ๋ผ๋ฏธํ„ฐ์ธ ์ •์ „์šฉ๋Ÿ‰ยท์ธ๋•ํ„ด์Šค ํ–‰๋ ฌ์„ ํ™œ์šฉํ•ด ํŒŒ๋™ ์‚ฐ๋ž€ ์ „์ด ํ–‰๋ ฌ์„ ๋„์ถœํ•จ์œผ๋กœ์จ, ์—ฐ์‡„ํ˜• ๊ตฌ์กฐ์˜ ์ „์ฒด ์‘๋‹ต์„ ๊ฐ„๋‹จํžˆ ๊ณ„์‚ฐํ•  ์ˆ˜ ์žˆ๋Š” ์ˆ˜ํ•™์  ํ”„๋ ˆ์ž„์›Œํฌ๋ฅผ ์ œ๊ณตํ•œ๋‹ค. ์ด๋Š” ์„ค๊ณ„ ๋‹จ๊ณ„์—์„œ ์‹œ๋ฎฌ๋ ˆ์ด์…˜๊ณผ ํ•ด์„์„ ์ผ์น˜์‹œํ‚ค๋Š” ๋ฐ ํฐ ๋„์›€์ด ๋˜๋ฉฐ, ํŠนํžˆ ๋‹ค๋‹จ DCโ€‘๋ธ”๋กœ์ปค์™€ ๊ฐ™์€

Electrical Engineering and Systems Science
Time and ensemble averaging in time series analysis

Time and ensemble averaging in time series analysis

๋ณธ ๋…ผ๋ฌธ์€ ์‹œ๊ฐ„ ์‹œ๊ณ„์—ด ๋ถ„์„์—์„œ ํ”ํžˆ ์‚ฌ์šฉ๋˜๋Š” โ€œ์‹œ๊ฐ„ ํ‰๊ท  ์ง‘ํ•ฉ ํ‰๊ท โ€์ด๋ผ๋Š” ์—๋ฅด๊ณ ๋”• ๊ฐ€์ •์„ ์ •๋ฐ€ํžˆ ๊ฒ€์ฆํ•œ๋‹ค๋Š” ์ ์—์„œ ํ•™๋ฌธ์ ยท์‹ค์šฉ์  ์˜๋ฏธ๊ฐ€ ํฌ๋‹ค. ๊ธฐ์กด ๋ฌธํ—Œ์—์„œ๋Š” ํ•˜๋‚˜์˜ ๊ธด ๊ธฐ๋ก์„ ์ผ์ • ๊ธธ์ด๋กœ ์ž˜๋ผ ์—ฌ๋Ÿฌ ๊ฐœ์˜ ์กฐ๊ฐ์„ ๋งŒ๋“  ํ˜„์ƒํ•™์  ์ง‘ํ•ฉ (phenomenological ensemble) โ€“ ์ฆ‰ STE โ€“ ์ด MTE์™€ ํ†ต๊ณ„์ ์œผ๋กœ ๋™์ผํ•˜๋‹ค๊ณ  ๊ฐ€์ •ํ•ด ์™”๋‹ค. ๊ทธ๋Ÿฌ๋‚˜ ์ €์ž๋“ค์€ ์ด ๊ฐ€์ •์ด ์‹ค์ œ ๋™์—ญํ•™์„ ์™œ๊ณกํ•  ๊ฐ€๋Šฅ์„ฑ์„ ์ง€์ ํ•˜๊ณ , ์ด๋ฅผ ๋ช…ํ™•ํžˆ ๊ตฌ๋ถ„ํ•˜๊ธฐ ์œ„ํ•ด OU ๋žญ๊ฒŒ์ธ ๋ฐฉ์ •์‹์ด๋ผ๋Š” ๊ฐ€์žฅ ๋‹จ์ˆœํ•˜๋ฉด์„œ๋„ ์ •์ƒยท์—๋ฅด๊ณ ๋”• ํŠน์„ฑ์„ ์™„๋ฒฝํžˆ ๊ฐ–์ถ˜ ๋ชจ๋ธ์„ ์„ ํƒํ•˜

Physics Analysis
Timed Parity Games: Complexity and Robustness

Timed Parity Games: Complexity and Robustness

๋ณธ ๋…ผ๋ฌธ์€ ์‹ค์‹œ๊ฐ„ ์‹œ์Šคํ…œ ์„ค๊ณ„ยท๊ฒ€์ฆ ๋ถ„์•ผ์—์„œ ํ•ต์‹ฌ์ ์ธ ๋‘ ๋ฌธ์ œ, ๋ณต์žก๋„ ๊ฐ์†Œ ์™€ ์ „๋žต์˜ ๊ฐ•์ธ์„ฑ ์„ ๋™์‹œ์— ๋‹ค๋ฃจ๊ณ  ์žˆ๋‹ค. ๋จผ์ €, ํƒ€์ž„๋“œ ์˜คํ† ๋งˆํ†ค ์œ„์—์„œ ๋™์‹œ์ ์œผ๋กœ ํ–‰๋™๊ณผ ์ง€์—ฐ์„ ์ œ์•ˆํ•˜๋Š” ๊ฒŒ์ž„์€ ๋ณธ์งˆ์ ์œผ๋กœ ๋ฌดํ•œ ์ƒํƒœยท๋ฌดํ•œ ์‹œ๊ฐ„ ๊ณต๊ฐ„์„ ๊ฐ–๋Š”๋‹ค. ๊ธฐ์กด ์—ฐ๊ตฌ์—์„œ๋Š” ์ด๋Ÿฌํ•œ ๊ฒŒ์ž„์„ ์ง์ ‘ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•ด ๋ณต์žก๋„๊ฐ€ EXPTIME ์ˆ˜์ค€์— ์ด๋ฅด๋Š” ์•Œ๊ณ ๋ฆฌ์ฆ˜์ด ์ œ์‹œ๋˜์—ˆ์œผ๋ฉฐ, ์‹ค์ œ ์ ์šฉ์— ํ•œ๊ณ„๊ฐ€ ์žˆ์—ˆ๋‹ค. ์ €์ž๋“ค์€ ์‹œ๊ณ„ ์˜์—ญ(clock regions) ๊ฐœ๋…์„ ํ™œ์šฉํ•ด ๋ฌดํ•œ ์ƒํƒœ ๊ณต๊ฐ„์„ ์œ ํ•œํ•œ ์˜์—ญ ๊ทธ๋ž˜ํ”„๋กœ ์ถ”์ƒํ™”ํ•œ๋‹ค๋Š” ์ ์—์„œ ๊ธฐ์กด ์ ‘๊ทผ๋ฒ•๊ณผ ์ฐจ๋ณ„ํ™”๋œ๋‹ค. ์ค‘์š”ํ•œ ์ 

Computer Science Logic
Toward Intelligent Autonomous Agents for Cyber Defense: Report of the   2017 Workshop by the North Atlantic Treaty Organization (NATO) Research Group   IST-152-RTG

Toward Intelligent Autonomous Agents for Cyber Defense: Report of the 2017 Workshop by the North Atlantic Treaty Organization (NATO) Research Group IST-152-RTG

์ด ๋ณด๊ณ ์„œ๋Š” NATO ์ฐจ์›์˜ ์‚ฌ์ด๋ฒ„ ๋ฐฉ์–ด ์ „๋žต์— ์ž์œจ ์—์ด์ „ํŠธ๋ฅผ ๋„์ž…ํ•˜๋ ค๋Š” ์ดˆ๊ธฐ ์‹œ๋„๋ผ ํ•  ์ˆ˜ ์žˆ๋‹ค. 2017๋…„ ๋‹น์‹œ ์‚ฌ์ด๋ฒ„ ์œ„ํ˜‘์€ ๊ธ‰๊ฒฉํžˆ ๊ณ ๋„ํ™”๋˜๊ณ  ์žˆ์—ˆ์œผ๋ฉฐ, ํŠนํžˆ ์•…์„ฑ์ฝ”๋“œ์™€ ์นจํˆฌ ๊ณต๊ฒฉ์ด ์ž๋™ํ™”ยท๋‹ค์ค‘ ๋‹จ๊ณ„๋กœ ์ „๊ฐœ๋˜๋Š” ์ถ”์„ธ์˜€๋‹ค. ๋ณด๊ณ ์„œ๋Š” ์ด๋Ÿฌํ•œ ์ƒํ™ฉ์„ โ€œ์นœํ™” ๋„คํŠธ์›Œํฌ๋ฅผ ์ง€์†์ ์œผ๋กœ ์ˆœ์ฐฐํ•˜๊ณ , ์ธ๊ฐ„์˜ ๋ฐ˜์‘ ์‹œ๊ฐ„๋ณด๋‹ค ๋น ๋ฅด๊ฒŒ ์ ๋Œ€์  ํ™œ๋™์„ ํƒ์ง€ยท๋Œ€์‘โ€ํ•ด์•ผ ํ•œ๋‹ค๋Š” ์‹œ๊ฐ„โ€‘์šฐ์œ„ ์›์น™์œผ๋กœ ์š”์•ฝํ•œ๋‹ค. 1. ์šด์˜ ํ™˜๊ฒฝ๊ณผ ๋ฐ์ดํ„ฐ ์ˆ˜์ง‘ ๋ณด๊ณ ์„œ๋Š” ์—์ด์ „ํŠธ๊ฐ€ ๋ฐฐ์น˜๋  ์ปดํ“จํŒ…ยท์ „์ˆ  ํ™˜๊ฒฝ ์„ ๋‘ ์ถ•์œผ๋กœ ๊ตฌ๋ถ„ํ•œ๋‹ค. ์ฒซ ๋ฒˆ์งธ๋Š” ์ „ํ†ต์ ์ธ IT ์ธํ”„๋ผ(์„œ๋ฒ„, ๋„คํŠธ์›Œํฌ

Computer Science Cryptography and Security
Towards Automatic & Personalised Mobile Health Interventions: An   Interactive Machine Learning Perspective

Towards Automatic & Personalised Mobile Health Interventions: An Interactive Machine Learning Perspective

๋ณธ ๋…ผ๋ฌธ์€ ๋ชจ๋ฐ”์ผ ํ—ฌ์Šค์ผ€์–ด ๋ถ„์•ผ์—์„œ โ€˜Humanโ€‘inโ€‘theโ€‘loopโ€™ ๋ฐฉ์‹์„ ํ™œ์šฉํ•œ ์ธํ„ฐ๋ž™ํ‹ฐ๋ธŒ ๋จธ์‹ ๋Ÿฌ๋‹(iML) ํ”„๋ ˆ์ž„์›Œํฌ๋ฅผ ์„ค๊ณ„ยท์‹œ์—ฐํ•จ์œผ๋กœ์จ, ๊ธฐ์กด์˜ ์ผ๋ฅ ์ ์ธ ๊ฑด๊ฐ• ๊ด€๋ฆฌ ํ”„๋กœ๊ทธ๋žจ์ด ๊ฐ–๋Š” ํ•œ๊ณ„๋ฅผ ๊ทน๋ณตํ•˜๊ณ ์ž ํ•œ๋‹ค. ์ฃผ์š” ๊ธฐ์—ฌ๋Š” ๋‹ค์Œ๊ณผ ๊ฐ™์ด ์ •๋ฆฌํ•  ์ˆ˜ ์žˆ๋‹ค. 1. ์‹œ์Šคํ…œ ์•„ํ‚คํ…์ฒ˜์™€ ๋ชจ๋“ˆํ™” ๋…ผ๋ฌธ์€ ๊ณ ์ˆ˜์ค€ ์•„ํ‚คํ…์ฒ˜๋ฅผ ์ œ์‹œํ•˜๋ฉด์„œ ๋ฐ์ดํ„ฐ ์ˆ˜์ง‘ยท์ „์ฒ˜๋ฆฌยท๋ชจ๋ธ ํ•™์Šตยทํ”ผ๋“œ๋ฐฑ ์ œ๊ณต์ด๋ผ๋Š” ๋„ค ๊ฐ€์ง€ ํ•ต์‹ฌ ๋ชจ๋“ˆ์„ ๊ตฌ๋ถ„ํ•œ๋‹ค. ์„ผ์„œ ๊ธฐ๋ฐ˜ ํ™œ๋™ ๋ฐ์ดํ„ฐ์™€ ์‚ฌ์šฉ์ž๊ฐ€ ์ง์ ‘ ์ž…๋ ฅํ•œ ์‹๋‹จยท์šด๋™ ๊ธฐ๋ก์„ ์‹ค์‹œ๊ฐ„์œผ๋กœ ํ†ตํ•ฉํ•˜๊ณ , ์ด๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ๊ฐœ์ธ ํ”„๋กœํ•„์„ ์ง€์†์ ์œผ๋กœ ์—…๋ฐ์ดํŠธ

HCI Learning Artificial Intelligence Computers and Society Computer Science
Towards Practical Implementation of Deep Random Secrecy

Towards Practical Implementation of Deep Random Secrecy

๋ณธ ๋…ผ๋ฌธ์€ ์•”ํ˜ธํ•™์—์„œ โ€œ๋ฌดํ•œํ•œ ๊ณ„์‚ฐยท์ €์žฅ ๋Šฅ๋ ฅ์„ ๊ฐ€์ง„ ์ ๋Œ€์žโ€๋ผ๋Š” ๊ฐ€์žฅ ๋ณด์ˆ˜์ ์ธ ๋ณด์•ˆ ๋ชจ๋ธ์„ ์ฑ„ํƒํ•จ์œผ๋กœ์จ, ๊ธฐ์กด์˜ ๋ณต์žก๋„ ๊ธฐ๋ฐ˜ ์•”ํ˜ธ ์ฒด๊ณ„๊ฐ€ ๊ฐ–๋Š” ๊ฐ€์ •(์˜ˆ: ์†Œ์ธ์ˆ˜๋ถ„ํ•ด ์–ด๋ ค์›€, ์ด์‚ฐ ๋กœ๊ทธ ๋‚œ์ด๋„)๊ณผ๋Š” ๊ทผ๋ณธ์ ์œผ๋กœ ๋‹ค๋ฅธ ์ ‘๊ทผ์„ ์‹œ๋„ํ•œ๋‹ค. ์ด๋Ÿฌํ•œ ๋ชจ๋ธ์€ ์ •๋ณดโ€‘์ด๋ก ์  ๋ณด์•ˆ์˜ ์ด์ƒํ˜•์— ๊ฐ€๊น์ง€๋งŒ, ์‹ค์ œ ๊ตฌํ˜„ ๊ฐ€๋Šฅ์„ฑ์— ๋Œ€ํ•œ ์˜๋ฌธ์ด ํ•ญ์ƒ ์ œ๊ธฐ๋œ๋‹ค. ์ €์ž๋“ค์€ ์ด๋ฅผ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•ด ๋‘ ๊ฐ€์ง€ ํ•ต์‹ฌ ๊ธฐ์—ฌ๋ฅผ ์ œ์‹œํ•œ๋‹ค. ์ฒซ์งธ, Cryptologic Limit ์ด๋ผ๋Š” ์ ˆ๋Œ€ ์ƒ์ˆ˜๋ฅผ ๋„์ž…ํ•ด DRS ํ”„๋กœํ† ์ฝœ์ด ๋‹ฌ์„ฑํ•  ์ˆ˜ ์žˆ๋Š” ๋น„๋ฐ€ ์ „์†ก๋ฅ ์˜ ์ด๋ก ์  ์ƒํ•œ์„ ๋ช…์‹œํ•œ๋‹ค. ์ด

Computer Science Cryptography and Security
Towards Theory of Massive-Parallel Proofs. Cellular Automata Approach

Towards Theory of Massive-Parallel Proofs. Cellular Automata Approach

์ด ๋…ผ๋ฌธ์€ ์ฆ๋ช… ์ด๋ก ์˜ ๊ทผ๋ณธ์ ์ธ ์ „์ œ๋ฅผ ๋’ค์ง‘๋Š” ์‹œ๋„๋ฅผ ํ•œ๋‹ค๋Š” ์ ์—์„œ ํ•™์ˆ ์  ์˜์˜๊ฐ€ ํฌ๋‹ค. ์ „ํ†ต์ ์ธ ํ˜•์‹ ๋…ผ๋ฆฌ์—์„œ๋Š” โ€˜๊ณต๋ฆฌ โ†’ ์ •๋ฆฌโ€™๋ผ๋Š” ์ผ๋ฐฉํ–ฅ ํ๋ฆ„์ด ํ•ต์‹ฌ์ด๋ฉฐ, ํŒŒ์ƒ ํŠธ๋ฆฌ๋Š” ์œ ํ•œํ•˜๊ณ  ๋น„์ˆœํ™˜์ ์ด์–ด์•ผ ํ•œ๋‹ค๋Š” ์ œ์•ฝ์ด ์žˆ๋‹ค. ์ €์ž๋Š” ์ด๋Ÿฌํ•œ ์ œ์•ฝ์„ ์™„์ „ํžˆ ์—†์• ๊ณ , ๋Œ€์‹  ์…€๋ฃฐ๋Ÿฌ ์˜คํ† ๋งˆํ†ค์ด๋ผ๋Š” ์ด์‚ฐ์  ๋™์—ญํ•™ ์‹œ์Šคํ…œ์— ์ฆ๋ช… ๊ณผ์ •์„ ๋งคํ•‘ํ•œ๋‹ค. ์—ฌ๊ธฐ์„œ ๊ฐ€์žฅ ํ˜์‹ ์ ์ธ ์š”์†Œ๋Š” ์ถ”๋ก  ๊ทœ์น™์„ ์ „์ด ํ•จ์ˆ˜ (delta) ๋กœ ํ•ด์„ ํ•œ๋‹ค๋Š” ์ ์ด๋‹ค. ์ „์ด ํ•จ์ˆ˜๋Š” ์ž…๋ ฅ(์…€ ์ž์ฒด์™€ ์ด์›ƒ ์…€๋“ค์˜ ์ƒํƒœ) โ†’ ์ถœ๋ ฅ(์ƒˆ๋กœ์šด ์ƒํƒœ)์ด๋ผ๋Š” ํ˜•ํƒœ๋ฅผ ๊ฐ–๋Š”๋ฐ, ์ด๋Š” ๋…ผ๋ฆฌ์  ํ•จ์˜(

Computer Science Logic Mathematics
Training a Large Scale Classifier with the Quantum Adiabatic Algorithm

Training a Large Scale Classifier with the Quantum Adiabatic Algorithm

๋ณธ ๋…ผ๋ฌธ์€ ๋ถ€์ŠคํŒ…(Boosting)์ด๋ผ๋Š” ๊ณ ์ „์ ์ธ ๊ธฐ๊ณ„ํ•™์Šต ํ”„๋ ˆ์ž„์›Œํฌ๋ฅผ ์–‘์ž ์ตœ์ ํ™”์™€ ์—ฐ๊ฒฐ์‹œํ‚จ ์ตœ์ดˆ์˜ ์‹œ๋„ ์ค‘ ํ•˜๋‚˜๋กœ ํ‰๊ฐ€ํ•  ์ˆ˜ ์žˆ๋‹ค. ๊ธฐ์กด AdaBoost๋Š” ์ง€์ˆ˜ ์†์‹ค ์„ ์ตœ์†Œํ™”ํ•˜๋Š” ํƒ์š•์ (greedy) ์•Œ๊ณ ๋ฆฌ์ฆ˜์ด๋ฉฐ, ๊ฐ€์ค‘์น˜๊ฐ€ ์–‘์˜ ์‹ค์ˆ˜๊ฐ’์„ ๊ฐ–๋Š”๋‹ค. ๋ฐ˜๋ฉด ์ €์ž๋“ค์€ ๋‘ ๊ฐ€์ง€ ํ•ต์‹ฌ์ ์ธ ๋ณ€ํ™”๋ฅผ ๋„์ž…ํ•œ๋‹ค. ์ฒซ์งธ, ์ด์ง„ ๊ฐ€์ค‘์น˜(0/1) ๋กœ ์ œํ•œํ•จ์œผ๋กœ์จ ์ตœ์ ํ™” ๋ณ€์ˆ˜๋ฅผ ์ด์‚ฐํ™”ํ•˜๊ณ , ๋‘˜์งธ, Lโ‚€โ€‘๋…ธ๋ฆ„ ์ •๊ทœํ™” ์™€ ์ด์ฐจ ์†์‹ค(leastโ€‘squares) ๋ฅผ ๊ฒฐํ•ฉํ•œ๋‹ค. Lโ‚€โ€‘๋…ธ๋ฆ„ ์ •๊ทœํ™”๋Š” ์‚ฌ์šฉ๋˜๋Š” ์•ฝํ•œ ๋ถ„๋ฅ˜๊ธฐ์˜ ์ˆ˜๋ฅผ ์ง์ ‘์ ์œผ๋กœ ์–ต์ œํ•œ๋‹ค. VC ์ฐจ์› ์ด

Computer Science Quantum Physics Machine Learning
Transport in networks with multiple sources and sinks

Transport in networks with multiple sources and sinks

์ด ๋…ผ๋ฌธ์€ ๋ณต์ˆ˜์˜ ์ถœ๋ฐœ์ ๊ณผ ๋„์ฐฉ์ ์ด ๋™์‹œ์— ํ™œ์„ฑํ™”๋˜๋Š” ์ƒํ™ฉ์„ ๋„คํŠธ์›Œํฌ ์ „์†ก ์ด๋ก ์— ์ •์‹์œผ๋กœ ๋„์ž…ํ•จ์œผ๋กœ์จ, ๊ธฐ์กด ์—ฐ๊ตฌ๊ฐ€ ์ฃผ๋กœ ๋‹ค๋ฃจ๋˜ ๋‹จ์ผ ์Œ ์ „์†ก ๋ชจ๋ธ์„ ํ™•์žฅํ•œ๋‹ค๋Š” ์ ์—์„œ ํ•™๋ฌธ์  ์˜์˜๊ฐ€ ํฌ๋‹ค. ์ €์ž๋“ค์€ ๋จผ์ € ์ „๋ฅ˜์™€ ํ๋ฆ„์ด๋ผ๋Š” ๋‘ ๊ฐ€์ง€ ์ „ํ˜•์ ์ธ ์ „์†ก ๋ฉ”์ปค๋‹ˆ์ฆ˜์„ ๊ตฌ๋ถ„ํ•œ๋‹ค. ์ „๋ฅ˜๋Š” ๊ฐ ๋งํฌ๊ฐ€ ๋‹จ์œ„ ์ €ํ•ญ์„ ๊ฐ–๋Š” ์ „๊ธฐ ํšŒ๋กœ ๋ชจ๋ธ์— ๊ธฐ๋ฐ˜ํ•ด ๊ฒฝ๋กœ ๊ธธ์ด์— ๋น„๋ก€ํ•œ ์†์‹ค์„ ํฌํ•จํ•˜๊ณ , ํ๋ฆ„์€ ๋ชจ๋“  ๋งํฌ๊ฐ€ ๋‹จ์œ„ ์šฉ๋Ÿ‰์„ ๊ฐ–๋Š” ์ตœ๋Œ€ ํ๋ฆ„(maxโ€‘flow) ๋ฌธ์ œ์™€ ๋™๋“ฑํ•˜๊ฒŒ ์ •์˜ํ•œ๋‹ค. ์ด ๋‘ ๋ชจ๋ธ์„ ๋™์ผํ•œ ๋„คํŠธ์›Œํฌ ๊ตฌ์กฐ์— ์ ์šฉํ•จ์œผ๋กœ์จ, โ€œ๋ณ‘๋ชฉ ํ˜„์ƒโ€์ด๋ผ๋Š” ๊ณตํ†ต๋œ

Condensed Matter Computer Science Discrete Mathematics Network
No Image

Two generalizations on the minimum Hamming distance of repeated-root constacyclic codes

์ด ์—ฐ๊ตฌ๋Š” ๋ฐ˜๋ณต๊ทผ(constacyclic) ์ฝ”๋“œ๋ฅผ ๋‹ค๋ฃจ๋Š” ๋ถ„์•ผ์—์„œ ์ค‘์š”ํ•œ ๋‘ ๊ฐ€์ง€ ํ™•์žฅ์„ ์ œ์‹œํ•œ๋‹ค. ์ฒซ ๋ฒˆ์งธ ํ™•์žฅ์€ ๊ธธ์ด (np^{s})์ธ ์ฝ”๋“œ๊ตฐ์— ๋Œ€ํ•œ ์ตœ์†Œ ํ•ด๋ฐ ๊ฑฐ๋ฆฌ ๊ณต์‹์ด๋‹ค. ๊ธฐ์กด ๋ฌธํ—Œ์—์„œ๋Š” ์ฃผ๋กœ ๊ธธ์ด (p^{s}) ํ˜น์€ (2p^{s})์ธ ์ˆœํ™˜ยท๋ถ€ํ˜ธ์ˆœํ™˜ ์ฝ”๋“œ์— ํ•œ์ •๋œ ๊ฒฐ๊ณผ๊ฐ€ ์ œ์‹œ๋˜์—ˆ์œผ๋ฉฐ, ๊ทธ ๊ทผ๊ฑฐ๋Š” ๋ณต์žกํ•œ ๊ฐ€์šฐ์Šค ํ•ฉ์ด๋‚˜ ์ฒด ํ™•์žฅ์˜ ์„ฑ์งˆ์„ ์ด์šฉํ•œ ๊ฒฝ์šฐ๊ฐ€ ๋งŽ์•˜๋‹ค. ์ €์ž๋Š” (x^{n}+gamma)๊ฐ€ ๊ธฐ์•ฝ๋‹คํ•ญ์‹์ด๋ผ๋Š” ๊ฐ€์ • ํ•˜์—, ์ƒ์„ฑ๋‹คํ•ญ์‹ ((x^{n}+gamma)^{ell})๊ฐ€ ์ •์˜ํ•˜๋Š” ์ฝ”๋“œ์˜ ๊ตฌ์กฐ๋ฅผ ์ฒด (

Computer Science Information Theory Mathematics
Two-dimensional generalization of the Muller root-finding algorithm and   its applications

Two-dimensional generalization of the Muller root-finding algorithm and its applications

: ํ˜„๋Œ€ ๋ฌผ๋ฆฌํ•™์—์„œ ๊ทผ ์ฐพ๊ธฐ ์•Œ๊ณ ๋ฆฌ์ฆ˜์€ Heun ํ•จ์ˆ˜์™€ ๊ฐ™์€ ๋ณต์žกํ•œ ํ•จ์ˆ˜๋ฅผ ์ฒ˜๋ฆฌํ•˜๋Š” ๋ฐ ํฐ ๋„์ „์„ ์•ˆ๊ฒจ์ค๋‹ˆ๋‹ค. Heun ํ•จ์ˆ˜๋Š” ์—ฌ๋Ÿฌ ํŠน์ด์ ์„ ๊ฐ–๋Š” ์ด์ฐจ ์„ ํ˜• ์ƒ๋ฏธ๋ถ„ ๋ฐฉ์ •์‹์˜ ๊ณ ์œ  ํ•ด๋กœ, ์ดˆ์›” ๊ธฐํ•˜ ํ•จ์ˆ˜์™€ ๋ฐ€์ ‘ํ•œ ๊ด€๋ จ์ด ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ํ•จ์ˆ˜์˜ ๋ถ„์„์  ์—ฐ๊ตฌ๋Š” ์•„์ง ๋ถˆ์™„์ „ํ•˜๋ฉฐ, ์ด๋ฅผ ์ฒ˜๋ฆฌํ•  ์ˆ˜ ์žˆ๋Š” ์†Œํ”„ํŠธ์›จ์–ด ํŒจํ‚ค์ง€๋Š” maple์ด ์œ ์ผํ•ฉ๋‹ˆ๋‹ค. ๊ทธ๋Ÿฌ๋‚˜ maple์˜ Heun ํ•จ์ˆ˜ ํ‰๊ฐ€ ๋ฃจํ‹ด์€ ํŠน์ • ๊ฒฝ์šฐ์—์„œ ์˜ค๋ฅ˜๋ฅผ ์ผ์œผํ‚ค๊ฑฐ๋‚˜ ์ •๋ฐ€๋„๊ฐ€ ๋–จ์–ด์งˆ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ํ•œ๊ณ„๋ฅผ ๊ทน๋ณตํ•˜๊ธฐ ์œ„ํ•ด ์ œ์•ˆ๋œ ์ƒˆ๋กœ์šด 2์ฐจ์› ๋ฎ๋Ÿฌ ์•Œ๊ณ ๋ฆฌ์ฆ˜์€ ๊ธฐ์กด ์ผ์ฐจ์› ์•Œ๊ณ ๋ฆฌ์ฆ˜

General Relativity Astrophysics Numerical Analysis HEP-TH Computer Science
Two-photon nonlinear spectroscopy of periodically trapped ultracold   atoms in a cavity

Two-photon nonlinear spectroscopy of periodically trapped ultracold atoms in a cavity

๋ณธ ๋…ผ๋ฌธ์€ โ€œ๋‘ ๊ด‘์ž ๋น„์„ ํ˜• ๋ถ„๊ด‘โ€์ด๋ผ๋Š” ๋น„๊ต์  ์ƒˆ๋กœ์šด ํƒ์ง€ ๊ธฐ๋ฒ•์„, ๊ด‘๊ฒฉ์ž์— ๊ฐ€๋‘์–ด์ง„ BEC์™€ ๊ณ โ€‘Q ์บ๋น„ํ‹ฐ์˜ ๋‘ ๋ชจ๋“œ ์‚ฌ์ด์— ์ ์šฉํ•จ์œผ๋กœ์จ ์ดˆ์ €์˜จ ์›์ž ๋ฌผ์งˆ์˜ ์–‘์ƒ ์ „์ด๋ฅผ ๋น„ํŒŒ๊ดด์ ์œผ๋กœ ๊ตฌ๋ถ„ํ•  ์ˆ˜ ์žˆ์Œ์„ ๋ณด์—ฌ์ค€๋‹ค. ๊ธฐ์กด์˜ ์ „ํ†ต์ ์ธ ๋ฐฉ๋ฒ•์€ ํŠธ๋žฉ์—์„œ ๋ฐฉ์ถœ๋œ ์›์ž ๊ตฌ๋ฆ„์˜ ๊ฐ„์„ญ ๋ฌด๋Šฌ๋ฅผ ๋ถ„์„ํ•˜๋Š” ํŒŒ๊ดด์  ์ธก์ •์— ์˜์กดํ–ˆ์œผ๋ฉฐ, ์ด๋Š” ์›์ž ์ˆ˜์™€ ์ƒ์ „์ด ์ •๋ณด๋ฅผ ๋™์‹œ์— ์–ป๊ธฐ ์–ด๋ ต๋‹ค๋Š” ํ•œ๊ณ„๊ฐ€ ์žˆ์—ˆ๋‹ค. ์—ฌ๊ธฐ์„œ๋Š” ์›์ž๋“ค์ด ํ˜•์„ฑํ•˜๋Š” ์–‘์ž ๊ตด์ ˆ๋ฅ ์ด ์บ๋น„ํ‹ฐ ์ „์†ก ์ŠคํŽ™ํŠธ๋Ÿผ์— ์ง์ ‘์ ์œผ๋กœ ๋งคํ•‘๋œ๋‹ค๋Š” ์ ์„ ์ด์šฉํ•œ๋‹ค. ๋‘ ๊ด‘์ž ์ „์ด๋Š” ์ค‘๊ฐ„ ๋ ˆ๋ฒจ |iโŸฉ๋ฅผ ๊ฐ€์ƒ์ ์œผ๋กœ ๊ฑฐ์ณ

Condensed Matter
Type-I bursts within outbursts of IGR J17473-2721

Type-I bursts within outbursts of IGR J17473-2721

: ๋ณธ ๋…ผ๋ฌธ์€ IGR J17473 2721์—์„œ ๊ด€์ฐฐ๋œ ์œ ํ˜• I X์„  ํญ๋ฐœ์„ ์‹ฌ์ธต์ ์œผ๋กœ ๋ถ„์„ํ•˜์—ฌ ์•„ํ†จ XRB์— ๋Œ€ํ•œ ์ƒˆ๋กœ์šด ์ดํ•ด๋ฅผ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค. ์ €์ž๋“ค์€ ๋จผ์ € ์œ ํ˜• I ํญ๋ฐœ์˜ ์ •์˜์™€ ํŠน์„ฑ์„ ์„ค๋ช…ํ•˜๋ฉฐ, ์ด๋Š” ์ค‘์„ฑ์ž๋ณ„ ํ‘œ๋ฉด์—์„œ ์ถ•์ ๋œ ๋ฌผ์งˆ์˜ ๋ถˆ์•ˆ์ •ํ•œ ์—ฐ์†Œ์— ๊ธฐ์ธํ•œ ๊ฒƒ์œผ๋กœ ์•Œ๋ ค์ ธ ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ํญ๋ฐœ์€ ์งง์€ ์ƒ์Šน ์‹œ๊ฐ„๊ณผ ๋А๋ฆฐ ๊ฐ์†Œ ์‹œ๊ฐ„์„ ๊ฐ€์ง€๋ฉฐ, ๋™๋ฐ˜ ๋ณ„๋กœ๋ถ€ํ„ฐ์˜ ๊ฐ•ํ•˜์œจ ์ฆ๊ฐ€์— ์˜ํ•ด ๋ฐœ์ƒํ•ฉ๋‹ˆ๋‹ค. IGR J17473 2721์€ ์•„ํ†จ ์†Œ์Šค๋กœ ๋ถ„๋ฅ˜๋˜๋ฉฐ, ๋‘ ๋ฒˆ์˜ ํญ๋ฐœ ๋™์•ˆ ๋‹ค์–‘ํ•œ ์ŠคํŽ™ํŠธ๋Ÿผ ์ƒํƒœ๋ฅผ ๋ณด์—ฌ์ฃผ์—ˆ์Šต๋‹ˆ๋‹ค. ์ €์ž๋“ค์€ RXTE์˜ ๊ด€์ธก ๋ฐ์ดํ„ฐ๋ฅผ

Astrophysics
Uncovering Competency Gaps in Large Language Models and Their Benchmarks

Uncovering Competency Gaps in Large Language Models and Their Benchmarks

[์ œ๋ชฉ KO] ์–ธ์–ด ๋ชจ๋ธ๊ณผ ๋ฒค์น˜๋งˆํฌ์˜ ์—ญ๋Ÿ‰ ๊ฒฉ์ฐจ ํ•ด๋ถ€ํ•˜๊ธฐ [์ดˆ๋ก KO] ๋Œ€๊ทœ๋ชจ ์–ธ์–ด ๋ชจ๋ธ(LLM)์˜ ํ‰๊ฐ€๋Š” ํ‘œ์ค€ํ™”๋œ ๋ฒค์น˜๋งˆํฌ์— ํฌ๊ฒŒ ์˜์กดํ•œ๋‹ค. ์ด๋Ÿฌํ•œ ๋ฒค์น˜๋งˆํฌ๋Š” ํŠน์ • ๋Šฅ๋ ฅ ์ธก๋ฉด์—์„œ ์œ ์šฉํ•œ ์ง‘๊ณ„๋œ ์ง€ํ‘œ๋ฅผ ์ œ๊ณตํ•˜์ง€๋งŒ, ์ด๋Ÿฌํ•œ ์ง‘๊ณ„๋œ ์ง€ํ‘œ๋Š” (i) LLM์ด ์•ฝํ•œ ํŠน์ • ํ•˜์œ„ ์˜์—ญ('๋ชจ๋ธ ๊ฒฉ์ฐจ')๊ณผ (ii) ๋ฒค์น˜๋งˆํฌ ์ž์ฒด์—์„œ ๊ท ํ˜•์ด ๋งž์ง€ ์•Š๋Š” ์ปค๋ฒ„๋ฆฌ์ง€('๋ฒค์น˜๋งˆํฌ ๊ฒฉ์ฐจ')๋ฅผ ๊ฐ€๋ฆด ์ˆ˜ ์žˆ๋‹ค. ์šฐ๋ฆฌ๋Š” ์ž๋™์œผ๋กœ ๋‘ ๊ฐ€์ง€ ์œ ํ˜•์˜ ๊ฒฉ์ฐจ๋ฅผ ๋ชจ๋‘ ์ฐพ์•„๋‚ด๋Š” ์ƒˆ๋กœ์šด ๋ฐฉ๋ฒ•์„ ์ œ์•ˆํ•œ๋‹ค. ์ด ๋ฐฉ๋ฒ•์€ ํฌ์†Œ ์ž๋™ ์ธ์ฝ”๋”(SAE)๋ฅผ ์‚ฌ์šฉํ•˜์—ฌ ๋ชจ๋ธ ๋‚ด๋ถ€ ํ‘œํ˜„์— ๊ธฐ๋ฐ˜์„ ๋‘

Model
No Image

Understanding the Generalization of Stochastic Gradient Adam in Learning Neural Networks

: ๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” ์Šคํ† ์บ์Šคํ‹ฑ ์•„๋‹ด์ด ์™œ ํ’€ ๋ฐฐ์น˜ ๋ฒ„์ „๊ณผ ๋‹ค๋ฅธ ์ˆ˜๋ ด์ ์„ ๊ฐ–๋Š”์ง€, ๊ทธ๋ฆฌ๊ณ  ๋ฏธ๋‹ˆ๋ฐฐ์น˜๊ฐ€ ์–ด๋–ป๊ฒŒ ๊ทผ์‚ฌ์ ์ธ ์ตœ์ ํ•ด๋ฅผ ์ฐพ๋Š” ๋ฐ ๋„์›€์„ ์ฃผ๋Š”์ง€์— ๋Œ€ํ•ด ์‹ฌ๋„ ์žˆ๊ฒŒ ๋ถ„์„ํ•ฉ๋‹ˆ๋‹ค. ๋˜ํ•œ, ๊ฐ€์ค‘์น˜ ๊ฐ์‡ ์˜ ์ค‘์š”์„ฑ๊ณผ ์•„๋‹ด ๋ฐ ์•„๋‹ดW ๊ฐ„์˜ ์ฐจ์ด์ ์— ๋Œ€ํ•œ ์ด๋ก ์  ๊ทผ๊ฑฐ๋ฅผ ์ œ์‹œํ•ฉ๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ๋ถ„์„์€ ์‹ค์ œ ๋”ฅ๋Ÿฌ๋‹ ๋ชจ๋ธ์„ ํ›ˆ๋ จํ•  ๋•Œ ๊ณ ๋ คํ•ด์•ผ ํ•  ์ค‘์š”ํ•œ ์š”์†Œ๋“ค์„ ์ œ๊ณตํ•˜๋ฉฐ, ํŠนํžˆ ๊ฐ€์ค‘์น˜ ๊ฐ์‡ ์™€ ๋ฐฐ์น˜ ํฌ๊ธฐ์˜ ์ ์ ˆํ•œ ์„ ํƒ์ด ์ผ๋ฐ˜ํ™” ์„ฑ๋Šฅ์— ๋ฏธ์น˜๋Š” ์˜ํ–ฅ์„ ๊ฐ•์กฐํ•ฉ๋‹ˆ๋‹ค.

Network Learning
No Image

Universality for the focusing nonlinear Schroedinger equation at the gradient catastrophe point: Rational breathers and poles of the tritronquee solution to Painleve I

: ์ง‘์ค‘ํ˜• ๋น„์„ ํ˜• ์Šˆ๋ขฐ๋”ฉ๊ฑฐ ๋ฐฉ์ •์‹์€ ๊ด‘์„ฌ์œ ์—์„œ์˜ ๋น„์„ ํ˜• ์ „์†ก๊ณผ ๊ฐ™์€ ๋‹ค์–‘ํ•œ ๋ฌผ๋ฆฌ์  ํ˜„์ƒ์„ ์„ค๋ช…ํ•˜๋Š” ์ค‘์š”ํ•œ ๋ชจ๋ธ์ž…๋‹ˆ๋‹ค. ์ด ๋…ผ๋ฌธ์—์„œ๋Š” NLS์˜ ์ค€๊ณ ์ „์  ๊ทนํ•œ์„ ๋ถ„์„ํ•˜์—ฌ ์กฐ์ ˆ ๋ถˆ์•ˆ์ •์„ฑ์—๋„ ๋ถˆ๊ตฌํ•˜๊ณ  ๋‚˜ํƒ€๋‚˜๋Š” ์งˆ์„œ ์žˆ๋Š” ๊ตฌ์กฐ๋ฅผ ์ดํ•ดํ•˜๊ณ ์ž ํ•ฉ๋‹ˆ๋‹ค. NLS ๋ฐฉ์ •์‹์˜ ์ดˆ๊ธฐ ๋ฐ์ดํ„ฐ๊ฐ€ ๋ณ€์กฐ ํ‰๋ฉดํŒŒ์˜ ํ˜•ํƒœ๋ฅผ ์ทจํ•  ๋•Œ, Forest์™€ Lee

MATH-PH Nonlinear Sciences Mathematics
Universally Utility-Maximizing Privacy Mechanisms

Universally Utility-Maximizing Privacy Mechanisms

: ์ด ๋…ผ๋ฌธ์€ ๋ฏผ๊ฐํ•œ ๋ฐ์ดํ„ฐ๊ฐ€ ํฌํ•จ๋œ ํ†ต๊ณ„ ๋ฐ์ดํ„ฐ๋ฒ ์ด์Šค๋ฅผ ๊ณต๊ฐœํ•˜๋Š” ๋ฉ”์ปค๋‹ˆ์ฆ˜์˜ ์ค‘์š”์„ฑ์„ ๊ฐ•์กฐํ•ฉ๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ๋ฉ”์ปค๋‹ˆ์ฆ˜์€ ๊ฐœ์ธ์ •๋ณด ๋ณดํ˜ธ์™€ ์œ ์šฉ์„ฑ ๊ฐ„์˜ ๊ท ํ˜•์„ ๋งž์ถ”๋Š” ๋ฐ ์žˆ์–ด ์ค‘์š”ํ•œ ์—ญํ• ์„ ํ•ฉ๋‹ˆ๋‹ค. ์ฐจ๋ถ„ ๊ฐœ์ธ์ •๋ณด ๋ณดํ˜ธ ํ”„๋ ˆ์ž„์›Œํฌ๋ฅผ ์‚ฌ์šฉํ•จ์œผ๋กœ์จ, ์—ฐ๊ตฌ์ž๋“ค์€ ๊ฐœ์ธ์ •๋ณด๋ฅผ ์—„๊ฒฉํ•˜๊ฒŒ ์ •๋Ÿ‰ํ™”ํ•˜๊ณ  ์ด๋ฅผ ๋ณด์žฅํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋Š” ๋ฐ์ดํ„ฐ๋ฒ ์ด์Šค ํ–‰์˜ ๋ณ€๊ฒฝ์— ๋”ฐ๋ผ ๋ฉ”์ปค๋‹ˆ์ฆ˜์˜ ์ถœ๋ ฅ ๋ถ„ํฌ๊ฐ€ ํฌ๊ฒŒ ๋‹ฌ๋ผ์ง€์ง€ ์•Š๋„๋ก ํ•˜๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค. ๋…ผ๋ฌธ์—์„œ ์ œ์•ˆํ•˜๋Š” ๊ธฐํ•˜ํ•™์  ๋ฉ”์ปค๋‹ˆ์ฆ˜ $M^ $์€ ๊ณ ์ •๋œ ๊ฐœ์ˆ˜ ์ฟผ๋ฆฌ์™€ ์ฐจ๋ถ„ ๊ฐœ์ธ์ •๋ณด ๋ณดํ˜ธ ์ˆ˜์ค€์— ๋Œ€ํ•œ ์ตœ์ ์˜ ์†”๋ฃจ์…˜์„ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค. ์ด ๋ฉ”

Computer Science Game Theory Databases
No Image

Unvalidated Trust: Cross-Stage Vulnerabilities in Large Language Model Architectures

๋ณธ ๋…ผ๋ฌธ์€ ๋Œ€ํ˜• ์–ธ์–ด ๋ชจ๋ธ์ด ๋‹ค์–‘ํ•œ ๋ถ„์•ผ์— ์ ์šฉ๋˜๋ฉด์„œ ๋ฐœ์ƒํ•˜๋Š” ์ž ์žฌ์  ์œ„ํ—˜์„ ์‹ฌ์ธต์ ์œผ๋กœ ๋ถ„์„ํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์—ฐ๊ตฌ์ž๋“ค์€ LLMs์˜ ๋‹ค๋‹จ๊ณ„ ์ฒ˜๋ฆฌ ๊ณผ์ •์—์„œ ์ž…๋ ฅ ๋ฐ์ดํ„ฐ๊ฐ€ ์ค‘๋ฆฝ์ ์œผ๋กœ ํ•ด์„๋˜์ง€ ์•Š๊ณ , ๊ตฌํ˜„๋œ ์•Œ๊ณ ๋ฆฌ์ฆ˜์— ๋”ฐ๋ผ ์™œ๊ณก๋˜๊ฑฐ๋‚˜ ํŽธํ–ฅ๋  ์ˆ˜ ์žˆ์Œ์„ ์ง€์ ํ•ฉ๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ๋น„์ค‘๋ฆฝ์ ์ธ ํ•ด์„์€ ์˜๋„ํ•˜์ง€ ์•Š์€ ์ƒํƒœ ๋ณ€ํ™”๋ฅผ ์œ ๋ฐœํ•˜๊ฑฐ๋‚˜, ์‚ฌ์šฉ์ž์˜ ๋ช…๋ น ์—†์ด๋„ ๋ชจ๋ธ์ด ์ž์œจ์ ์œผ๋กœ ์ž‘๋™ํ•˜๊ฒŒ ๋งŒ๋“ค ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋Š” ์•„ํ‚คํ…์ฒ˜์˜ ์‹คํŒจ ๋ชจ๋“œ๋ผ๊ณ  ์ •์˜ํ•˜๋ฉฐ, ๋‹จ์ˆœํ•œ ๋ฌธ์ž์—ด ํ•„ํ„ฐ๋ง๋งŒ์œผ๋กœ๋Š” ์ด๋Ÿฌํ•œ ๋ฌธ์ œ๋ฅผ ํ•ด๊ฒฐํ•  ์ˆ˜ ์—†๋‹ค๊ณ  ์ฃผ์žฅํ•ฉ๋‹ˆ๋‹ค. ๋…ผ๋ฌธ์—์„œ๋Š” 41๊ฐœ์˜ ์ƒ์šฉ LLMs์—

Model
Using Humanoid Robot to Instruct and Evaluate Performance of a Physical   Task

Using Humanoid Robot to Instruct and Evaluate Performance of a Physical Task

๋ณธ ๋…ผ๋ฌธ์€ ์‚ฌํšŒ์  ์ง€์› ๋กœ๋ด‡(Socially Assistive Robots) ๋ถ„์•ผ์—์„œ โ€˜์ง€์‹œยท๊ด€์ฐฐยทํ”ผ๋“œ๋ฐฑโ€™ ์‚ผ์œ„์ผ์ฒด ๊ตฌ์กฐ๋ฅผ ์‹ค์ œ ๋ฌผ๋ฆฌ ๊ณผ์ œ์— ์ ์šฉํ•œ ์ตœ์ดˆ ์‚ฌ๋ก€ ์ค‘ ํ•˜๋‚˜๋กœ ํ‰๊ฐ€ํ•  ์ˆ˜ ์žˆ๋‹ค. ๊ธฐ์กด ์—ฐ๊ตฌ๋“ค์€ ์ฃผ๋กœ ๋กœ๋ด‡์ด ๊ต์œกยท์žฌํ™œยท๋™๊ธฐ ๋ถ€์—ฌ ์—ญํ• ์„ ์ˆ˜ํ–‰ํ•˜๋Š” ๋ฐ ์ดˆ์ ์„ ๋งž์ถ”์—ˆ์œผ๋ฉฐ, ์‚ฌ์šฉ์ž์˜ ์ˆ˜ํ–‰ ๋Šฅ๋ ฅ์„ ์ •๋Ÿ‰์ ์œผ๋กœ ์ธก์ •ํ•˜๋Š” ๋ฐฉ๋ฒ•๋ก ์€ ์ƒ๋Œ€์ ์œผ๋กœ ๋ถ€์กฑํ–ˆ๋‹ค. ์ด ์ ์—์„œ ์ €์ž๋“ค์€ Boxโ€‘andโ€‘Blocks Test๋ผ๋Š” ํ‘œ์ค€ํ™”๋œ ์‹ ๊ฒฝ์‹ฌ๋ฆฌ ๊ฒ€์‚ฌ ๋„๊ตฌ๋ฅผ ๋ณ€ํ˜•ํ•จ์œผ๋กœ์จ, ๋กœ๋ด‡์ด ์ธ๊ฐ„ ๊ต์‚ฌ์™€ ๋™์ผํ•œ ์—ญํ• ์„ ์ˆ˜ํ–‰ํ•˜๋ฉด์„œ๋„ ์ž๋™ํ™”๋œ ๋ฐ์ดํ„ฐ ์ˆ˜์ง‘์ด ๊ฐ€๋Šฅํ•˜๋„๋ก ์„ค๊ณ„ํ•˜

Computer Science HCI
Variable TeV emission as a manifestation of jet formation in M87?

Variable TeV emission as a manifestation of jet formation in M87?

๋ณธ ๋…ผ๋ฌธ์€ M87์˜ ์ดˆ๊ณ ์—๋„ˆ์ง€(TeV) ๋ณ€๋™์ด ๋ธ”๋ž™ํ™€ ๋ฐ”๋กœ ์ฃผ๋ณ€์˜ ๋งˆ๊ทธ๋„คํ† ์Šคํ”ผ์–ด์—์„œ ๋ฐœ์ƒํ•œ๋‹ค๋Š” ๊ฐ€์„ค์„ ์ •๋Ÿ‰์ ์œผ๋กœ ๊ฒ€์ฆํ•œ๋‹ค. ํ•ต์‹ฌ ์•„์ด๋””์–ด๋Š” โ€˜๊ธฐ๋ฆ„์ง„โ€™ ๋งˆ๊ทธ๋„คํ† ์ŠคํŽ˜์–ด, ์ฆ‰ ๋ณต์‚ฌ ํšจ์œจ์ด ๋‚ฎ์€ RIAF(๋ฐฉ์‚ฌ ํšจ์œจ์ด ์–ต์ œ๋œ ํก์ˆ˜ ํ๋ฆ„)์—์„œ ๋ฐฉ์ถœ๋˜๋Š” MeV ๊ด‘์ž๊ฐ€ ๊ทน์ง€์˜ ์—ด๋ฆฐ ์ž๊ธฐ์žฅ ์„ ์„ ๋”ฐ๋ผ ์ „ํ•˜โ€‘์Œ์„ ์ƒ์„ฑํ•˜๊ณ , ์ด ์ „ํ•˜๊ฐ€ ์ง„๊ณต ๊ฐญ(vacuum gap) ๋‚ด๋ถ€์— ๊ณต๊ธ‰๋˜์–ด ์ „๊ธฐ์žฅ์— ์˜ํ•ด ์ž…์ž๋ฅผ ๊ฐ€์†ํ•œ๋‹ค๋Š” ๊ฒƒ์ด๋‹ค. 1. ์ „ํ•˜ ์ฃผ์ž… ๋ฉ”์ปค๋‹ˆ์ฆ˜ RIAF์—์„œ ๋ฐœ์ƒํ•œ MeV ๊ด‘์ž๋Š” (gammagamma) ์†Œ๋ฉธ์„ ํ†ตํ•ด ์ „์žยท์–‘์ „์ž ์Œ์„ ๋งŒ๋“ ๋‹ค. ์ „

Astrophysics
No Image

VFocus: Better Verilog Generation from Large Language Model via Focused Reasoning

์‚ฌ์ „ ์ˆœ์œ„ ์ง€์ • ๋‹จ๊ณ„ : VFocus๋Š” LLM ํ”„๋กฌํ”„ํŠธ๋ฅผ ํ†ตํ•ด ์—ฌ๋Ÿฌ ๊ฐœ์˜ ์ฝ”๋“œ ํ›„๋ณด๋ฅผ ์ƒ์„ฑํ•˜๊ณ , ๊ตฌ๋ฌธ์ ์œผ๋กœ ์œ ํšจํ•œ ์ถœ๋ ฅ์„ ์–ป๊ธฐ ์œ„ํ•ด ์žฌ์‹œ๋„ํ•˜๋ฉฐ, ๊ธฐ๋Šฅ์  ์ •ํ™•์„ฑ์„ ์œ„ํ•œ '์ถ”๋ก  ๋‹ฌ์ฝคํ•œ ์ง€์ ' ๋‚ด์— ์žˆ๋Š” ํ›„๋ณด๋ฅผ ์œ ์ง€ํ•˜๊ธฐ ์œ„ํ•ด ๋ฐ€๋„ ์œ ๋„ ํ•„ํ„ฐ๋ง ์„ ๋„์ž…ํ•ฉ๋‹ˆ๋‹ค. ์ˆœ์œ„ ์ง€์ • ๋‹จ๊ณ„ : ๊ฐ ์ฝ”๋“œ ํ›„๋ณด์— ๋Œ€ํ•ด ์ž๋™ ์ƒ์„ฑ๋œ ํ…Œ์ŠคํŠธ ๋ฒค์น˜๋ฅผ ์‚ฌ์šฉํ•˜์—ฌ ์‹œ๋ฎฌ๋ ˆ์ด์…˜์„ ์ˆ˜ํ–‰ํ•˜๊ณ , ์ž๊ธฐ ์ผ๊ด€์„ฑ ๊ธฐ๋ฐ˜ ํด๋Ÿฌ์Šคํ„ฐ๋ง์„ ์ ์šฉํ•˜์—ฌ ๊ฐ€์žฅ ์ผ๊ด€๋œ ์ถœ๋ ฅ์„ ์‹๋ณ„ํ•ฉ๋‹ˆ๋‹ค. ์‚ฌํ›„ ์ˆœ์œ„ ์ง€์ • ์ •์ œ ๋‹จ๊ณ„ : VFocus๋Š” ์ƒ์œ„ ๋žญํฌ๋œ ํ›„๋ณด์— ๋Œ€ํ•ด ๋ถˆ์ผ์น˜ ๋งˆ์ด๋‹์„ ์ˆ˜ํ–‰ํ•˜๊ณ , ์ถ”๋ก  ๊ฐ•ํ™” LL

Model
No Image

VolSegGS: Segmentation and Tracking in Dynamic Volumetric Scenes via Deformable 3D Gaussians

๋™์  ๋ณผ๋ฅจ ์žฅ๋ฉด์˜ ์„ธ๊ทธ๋จผํŠธ์™€ ์ถ”์ ์€ ๋ณต์žกํ•œ ํ˜„์ƒ์„ ์ดํ•ดํ•˜๊ณ  ๋ถ„์„ํ•˜๋Š” ๋ฐ ์žˆ์–ด ์ค‘์š”ํ•œ ์š”์†Œ์ž…๋‹ˆ๋‹ค. ํŠนํžˆ, ๋Œ€๊ทœ๋ชจ ์‹œ๋ฎฌ๋ ˆ์ด์…˜ ๋ฐ์ดํ„ฐ์˜ ๊ฒฝ์šฐ, ํšจ์œจ์ ์ธ ์‹œ๊ฐํ™”์™€ ํƒ์ƒ‰์€ ํ•„์ˆ˜์ ์ด๋ฉฐ, ์ด๋ฅผ ์œ„ํ•œ ์ปดํ“จํŒ… ์ž์›์˜ ์ตœ์ ํ™”๊ฐ€ ํ•„์š”ํ•ฉ๋‹ˆ๋‹ค. VolSegGS๋Š” ์ด๋Ÿฌํ•œ ๋ฌธ์ œ๋ฅผ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•ด ์‹ ๊ฒฝ๋ง ๊ธฐ๋ฐ˜์˜ ๊ฐ€์šฐ์‹œ์•ˆ ๋ถ„์‚ฐ ๊ธฐ๋ฒ•์„ ์ œ์•ˆํ•ฉ๋‹ˆ๋‹ค. ์ด ๋ฐฉ๋ฒ•์€ ์‹ค์‹œ๊ฐ„์œผ๋กœ ์ƒˆ๋กœ์šด ๋ทฐ๋ฅผ ํ•ฉ์„ฑํ•˜๊ณ , ์„ธ๊ทธ๋จผํŠธ์™€ ์ถ”์ ์„ ์ˆ˜ํ–‰ํ•˜์—ฌ ํƒ์ƒ‰์  ์‹œ๊ฐํ™”๋ฅผ ๊ฐ€๋Šฅํ•˜๊ฒŒ ํ•ฉ๋‹ˆ๋‹ค. ํŠนํžˆ, ์ €ํฌ์˜ ์ ‘๊ทผ๋ฒ•์€ ๋‹ค์Œ๊ณผ ๊ฐ™์€ ์žฅ์ ์„ ๊ฐ€์ง€๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค: 1. ๋ณ€ํ˜• ๊ฐ€๋Šฅํ•œ 3D ๊ฐ€์šฐ์‹œ์•ˆ: ๋™์  ๋ณผ๋ฅจ ์žฅ๋ฉด์„ ํ‘œ

When Cultures Meet: Modelling Cross-Cultural Knowledge Spaces

When Cultures Meet: Modelling Cross-Cultural Knowledge Spaces

๋ณธ ๋…ผ๋ฌธ์€ ๋‹ค๋ฌธํ™” ํ˜‘์—…์ด ์ผ์ƒํ™”๋œ ํ˜„๋Œ€ ์—ฐ๊ตฌยท๋น„์ฆˆ๋‹ˆ์Šค ํ™˜๊ฒฝ์„ ๋ฐฐ๊ฒฝ์œผ๋กœ, ๋ฌธํ™”์  ์ฐจ์ด๋ฅผ ์ฒด๊ณ„์ ์œผ๋กœ ๊ด€๋ฆฌํ•  ์ˆ˜ ์žˆ๋Š” ์ •๋ณด ์‹œ์Šคํ…œ ์„ค๊ณ„๋ผ๋Š” ์‹ค์šฉ์  ๊ณผ์ œ์— ์ ‘๊ทผํ•œ๋‹ค. ๊ฐ€์žฅ ํฐ ๊ฐ•์ ์€ โ€˜๋ฌธํ™” ์˜จํ†จ๋กœ์ง€โ€™๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ํ•œ ๊ตฌ์กฐ์  ์ ‘๊ทผ์ด๋‹ค. ์˜จํ†จ๋กœ์ง€๋Š” ๊ฐœ๋…ยท๊ด€๊ณ„ยท์†์„ฑ์„ ๋ช…์‹œ์ ์œผ๋กœ ์ •์˜ํ•จ์œผ๋กœ์จ, ์„œ๋กœ ๋‹ค๋ฅธ ๋ฌธํ™”๊ถŒ ์‚ฌ์šฉ์ž๊ฐ€ ๋™์ผํ•œ ๋ฐ์ดํ„ฐ ๋ชจ๋ธ ์œ„์—์„œ ์˜์‚ฌ์†Œํ†ตํ•  ์ˆ˜ ์žˆ๊ฒŒ ํ•œ๋‹ค. ์ด๋Š” ๊ธฐ์กด์˜ ๋‹จ์ˆœ ๋ฒˆ์—ญยท๊ฐ€์ด๋“œ ์ˆ˜์ค€์„ ๋„˜์–ด, ๋ฌธํ™”์  ์˜๋ฏธ์™€ ํ–‰๋™ ๊ทœ๋ฒ”์„ ๊ธฐ๊ณ„๊ฐ€ ์ดํ•ดํ•˜๊ณ  ์ง€์›ํ•˜๋„๋ก ๋งŒ๋“ ๋‹ค. ์‹œ์Šคํ…œ ๊ตฌํ˜„์— XML์„ ์„ ํƒํ•œ ์ด์œ ๋Š” ์ƒํ˜ธ ์šด์šฉ์„ฑ๊ณผ ํ™•์žฅ์„ฑ์„ ํ™•๋ณดํ•˜๊ธฐ

Model Computer Science HCI
Window-Based Greedy Contention Management for Transactional Memory

Window-Based Greedy Contention Management for Transactional Memory

: ๋ฉ€ํ‹ฐ์ฝ”์–ด ์•„ํ‚คํ…์ฒ˜์˜ ๋“ฑ์žฅ์œผ๋กœ ๋‹ค์ค‘ ์Šค๋ ˆ๋“œ ์†Œํ”„ํŠธ์›จ์–ด๊ฐ€ ๋„๋ฆฌ ์‚ฌ์šฉ๋˜๋ฉด์„œ ๋™์‹œ์„ฑ ๊ด€๋ จ ๋ฌธ์ œ, ํŠนํžˆ ๋ฉ”๋ชจ๋ฆฌ ์ ‘๊ทผ ์ถฉ๋Œ์ด ์ค‘์š”ํ•œ ์ด์Šˆ๋กœ ๋– ์˜ค๋ฅด๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ๋ฌธ์ œ๋ฅผ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•ด ์†Œํ”„ํŠธ์›จ์–ด ํŠธ๋žœ์žญ์…”๋„ ๋ฉ”๋ชจ๋ฆฌ(STM) ์‹œ์Šคํ…œ์ด ์ฃผ๋ชฉ๋ฐ›๊ณ  ์žˆ์œผ๋ฉฐ, ๋ณธ ๋…ผ๋ฌธ์—์„œ๋Š” STM ์‹œ์Šคํ…œ์˜ ํ•ต์‹ฌ์ธ ์ปจํ…์…˜ ๊ด€๋ฆฌ ์ „๋žต์„ ๊ฐœ์„ ํ•˜๋Š” ๋ฐ ์ดˆ์ ์„ ๋งž์ถ”๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ๊ธฐ์กด์˜ ํƒ์š•์Šค๋Ÿฌ์šด ์ปจํ…์…˜ ๋งค๋‹ˆ์ €๋Š” ๊ฐ ์Šค๋ ˆ๋“œ๋‹น ๋‹จ์ผ ํŠธ๋žœ์žญ์…˜๋งŒ ๊ณ ๋ คํ•˜์—ฌ, ๋ฆฌ์†Œ์Šค ์ˆ˜๊ฐ€ ์ฆ๊ฐ€ํ•จ์— ๋”ฐ๋ผ ์„ฑ๋Šฅ์ด ์„ ํ˜•์ ์œผ๋กœ ์ €ํ•˜๋˜๋Š” ํ•œ๊ณ„๋ฅผ ๊ฐ€์ง€๊ณ  ์žˆ์—ˆ์Šต๋‹ˆ๋‹ค. ์ด๋Ÿฌํ•œ ๋ฌธ์ œ๋ฅผ ํ•ด๊ฒฐํ•˜๊ธฐ ์œ„ํ•ด ๋ณธ ์—ฐ๊ตฌ์—์„œ๋Š”

Computer Science Data Structures Performance Distributed Computing

< Category Statistics (Total: 2829) >

Electrical Engineering and Systems Science
100
General
731
General Relativity
22
HEP-EX
17
HEP-LAT
3
HEP-PH
39
HEP-TH
19
MATH-PH
36
NUCL-EX
2
NUCL-TH
5
Quantum Physics
41

Start searching

Enter keywords to search articles

โ†‘โ†“
โ†ต
ESC
โŒ˜K Shortcut