4,554 research outputs found
Learning needs assessment examining current infection prevention and control knowledge, readiness, and training preferences among healthcare providers
No full textNicholas Ida, MPH; Judith A. Guzman Cottrill, DO; Roza Tammer, MPH, CIC; Rebecca Pierce, PhD, MS, BSN; Dat Tran, MD, MS.This archived document is maintained by the State Library of Oregon as part of the Oregon Documents Depository Program. It is for informational purposes and may not be suitable for legal purposes.Mode of access: Internet from the Oregon Government Publications Collection.Text in English
Knapsack based optimal policies for budget-limited multi-armed bandits
No full textIn budget-limited multi-armed bandit (MAB) problems, the learner’s actions are costly and constrained by a fixed budget. Consequently, an optimal exploitation policy may not be to pull the optimal arm repeatedly, as is the case in other variants of MAB, but rather to pull the sequence of different arms that maximises the agent’s total reward within the budget. This difference from existing MABs means that new approaches to maximising the total reward are required. Given this, we develop two pulling policies, namely: (i) KUBE; and (ii) fractional KUBE. Whereas the former provides better performance up to 40% in our experimental settings, the latter is computationally less expensive. We also prove logarithmic upper bounds for the regret of both policies, and show that these bounds are asymptotically optimal (i.e. they only differ from the best possible regret by a constant factor)
Phuc Tran: Author, Teacher, Tattoo Artist
No full textProfile of Phuc Tran. Tran taught classical languages in middle and high schools in Maine and New York for years. He also owns a tattoo shop in Portland and has written a highly acclaimed novel titled Sigh, Gone about his family\u27s move to Pennsylvania from Vietnam in 1975. In this piece, Tran briefly discusses community, language, and how it feels being an Asian American in Maine
TRAN Quyet Thang
No full text학위논문(박사)--아주대학교 일반대학원 :전자공학과,2017. 2CHAPTER 1: INTRODUCTION 1
1.1 Introduction to Light Modulating Devices 1
1.2 Introduction to Graphene 1
1.3 Overview of the Numerical Methods 3
1.4 Organization of Dissertation 7
CHAPTER 2: TUNABLE WIDE-ANGLE TUNNELING IN GRAPHENE-ASSISTED FRUSTRATED TOTAL INTERNAL REFLECTION 9
2.1 Introduction 9
2.2 Frustrated Total Internal Reflection (FTIR) Effect 10
2.3 Controllable Wide Angle Tunneling Effect with ENZ Effect in FTIR Configuration 11
2.4 Waveguide-type optical modulator based on a GA-FTIR structure 20
2.5 Chapter Summary 26
CHAPTER 3: LOW LOSS ELECTRICALLY CONTROLLABLE VERTICALLY COUPLED DIRECTIONAL COUPLER 28
3.1 Introduction 28
3.2 Vertically Coupled Directional Coupler 29
3.3 Chapter Summary 40
CHAPTER 4: OPTICAL PHASE MODULATOR BASED ON GRAPHENE EMBEDDED ALL PASS FILTER 42
4.1 Introduction 42
4.2 Phase Modulating All Pass Filter with Graphene 42
4.3 Chapter Summary 48
CHAPTER 5: COUPLED MODE THEORY OF PERFECT GRAPHENE ABSORBERS IN DUAL-MODE/SINGLE-MODE COUPLED RESONATORS SYSTEM 50
5.1 Introduction 50
5.2 Dual Mode - Single Mode Coupled Resonators System 51
5.3 Temporal Coupled Mode Theory of the "Triple-mode absorber" 52
5.4 Numerical verification of the CMT 59
5.5 Chapter Summary 66
CHAPTER 6: CONCLUSION AND FUTURE WORK 68
6.1 Conclusion 68
6.2 Future Work 68
REFERENCES 70
APPENDIX: AUTHOR'S PUBLICATIONS LIST 79DoctoralOne of the most important component of integrated photonics is a class of devices known as light modulation devices, which allow us to modulate and manipulate the flow of light, similar to the role transistor played in electronics. Only recently introduced, but graphene have shown incredible promises as a "miracle" material in electronics, with properties ranging from zero band gap, very high electrical mobility, ultra broadband optical responses, and the ability to drastically modify its optical properties through chemical or electrical doping.
In this dissertation, the author presented several unique nanostructures to exploit the aforementioned graphene characteristics to create light modulation devices with superior performance characteristics. Novel effects including wide angle extraordinary reflection causes by epsilon near zero effect and wide angle extraordinary transmission causes by coupling of plasmonic supermodes, phase modulation with near unity amplitude transmission with all pass filter, and graphene perfect absorber with a coupled system of dual mode/single mode resonator was thoroughly investigated, theoretically and numerically. The effects was also presented in practical nanostructures better suited for applications, which are also numerically investigated with various numerical methods
On the efficiency of data collection for multiple Naïve Bayes classifiers
No full textMany classification problems are solved by aggregating the output of a group of distinct predictors. In this respect, a popular choice is to assume independence and employ a Naïve Bayes classifier. When we have not just one but multiple classification problems at the same time, the question of how to assign the limited pool of available predictors to the individual classification problems arises. Empirical studies show that the policies we use to perform such assignments have a strong impact on the accuracy of the system. However, to date there is little theoretical understanding of this phenomenon. To help rectify this, in this paper we provide the first theoretical explanation of the accuracy gap between the most popular policies: the non-adaptive uniform allocation, and the adaptive allocation schemes based on uncertainty sampling and information gain maximisation. To do so, we propose a novel representation of the data collection process in terms of random walks. Then, we use this tool to derive new lower and upper bounds on the accuracy of the policies. These bounds reveal that the tradeoff between the number of available predictors and the accuracy has a different exponential rate depending on the policy used. By comparing them, we are able to quantify the advantage that the two adaptive policies have over the non-adaptive one for the first time, and prove that the probability of error of the former decays at more than double the exponential rate of the latter. Furthermore, we show in our analysis that this result holds both in the case where we know the accuracy of each individual predictor, and in the case where we only have access to a noisy estimate of it
Streaming Bayesian inference for crowdsourced classification
No full textA key challenge in crowdsourcing is inferring the ground truth from noisy and unreliable data. To do so, existing approaches rely on collecting redundant information from the crowd, and aggregating it with some probabilistic method. However, oftentimes such methods are computationally inefficient, are restricted to some specific settings, or lack theoretical guarantees. In this paper, we revisit the problem of binary classification from crowdsourced data. Specifically we propose Streaming Bayesian Inference for Crowdsourcing (SBIC), a new algorithm that does not suffer from any of these limitations. First, SBIC has low complexity and can be used in a real-time online setting. Second, SBIC has the same accuracy as the best state-of-the-art algorithms in all settings. Third, SBIC has provable asymptotic guarantees both in the online and offline settings. <br/
Crowdsourcing complex workflows under budget constraints
No full textWe consider the problem of task allocation in crowdsourcing systems with multiple complex workflows, each of which consists of a set of interdependent micro-tasks. We propose Budgeteer, an algorithm to solve this problem under a budget constraint. In particular, our algorithm first calculates an efficient way to allocate budget to each workflow. It then determines the number of inter-dependent micro-tasks and the price to pay for each task within each workflow, given the corresponding budget constraints. We empirically evaluate it on a well-known crowdsourcing-based text correction workflow using Amazon Mechanical Turk, and show that Budgeteer can achieve similar levels of accuracy to current benchmarks, but is on average 45% cheaper
Monologue, Dialogue, and Tran Vietnam
No full textA manuscript comprised of materials completed by O. W. Wolters before his death.O. W. Wolters was a twentieth-century historian of early Southeast Asia
who began his academic career with the study of early commercial
relations in the Malay world and the maritime empire of Srivijaya, which
dominated the Straits of Malacca and neighboring seas for several
centuries. During the last twenty-five years of his life, he became
interested in the Tran dynasty of Vietnam (1225-1400). From 1976 to 1996,
he published twelve articles about the Tran dynasty. When he died in 2000,
he left a nearly-completed manuscript of a book-length work about that
dynasty, which is herewith made available to the world of readers.
What makes this manuscript particularly interesting is how the author
shaped a work of historical research into what he liked to call a novel.
He became convinced that there was a certain way of thinking and speaking
that was distinctive to educated people in the Tran period, and he
believed that the best way to present this was through conversational
dialogue. He further presents the Tran way of thinking as a critical
perspective on the regimes that followed.
This manuscript also contains self-reflexive meditations on what the
author was endeavoring to achieve and his critique of his success in doing
so. He offers readers a rare glimpse into the craftsmanship of the most
creative and adventurous scholar of early Southeast Asia in his
generation
On the efficiency of data collection for crowdsourced classification
No full textThe quality of crowdsourced data is often highly variable. For this reason, it is common to collect redundant data and use statistical methods to aggregate it. Empirical studies show that the policies we use to collect such data have a strong impact on the accuracy of the system. However, there is little theoretical understanding of this phenomenon. In this paper we provide the first theoretical explanation of the accuracy gap between the most popular collection policies: the non-adaptive uniform allocation, and the adaptive uncertainty sampling and information gain maximisation. To do so, we propose a novel representation of the collection process in terms of random walks. Then, we use this tool to derive lower and upper bounds on the accuracy of the policies. With these bounds, we are able to quantify the advantage that the two adaptive policies have over the non-adaptive one for the first time
Phallostethus cuulong Shibukawa, Tran & Tran, 2012, new species
No full textPhallostethus cuulong, new species New Vietnamese name: Cá bụng đầu (Figures 1–4) Holotype. ZRC 53233, male, 24.2 mm SL, a branch of Hau River (a distributary of Mekong), Cu Lao Dung, Soc Trang Province, Vietnam (9 ° 30.8 ’ N, 106 ° 13.7 ’ E), 0–0.5 m depth, 31 July 2009, collected by K. Shibukawa. Paratypes. Total eight specimens (five males and three females), 20.0– 24.5 mm SL: CTU-P 2327, 1 specimen (female), 23.7 mm SL, Duyen Hai, Tra Vinh Province, Vietnam (9 ° 40.9 ’ N, 106 ° 30.7 ’ E), 0.3–0.8 m depths, 5 April 2009, collected by K. Shibukawa, V.V. Tran and L.X. Tran; CTU-P 2494, 1 specimen (male), 22.5 mm SL, My Thanh River (a distributary of Mekong), Vinh Chau, Soc Trang Province, Vietnam (9 ° 22.7 ’ N, 106 ° 0.7 ’ E), 0.5–1.2 m depths, 1 August 2009, collected by H.P. Ha, V.V. Tran and L.X. Tran; CTU-P 5020, 1 specimen (male, cleared and stained), 23.5 mm SL, Cho Lach, Ben Tre, Vietnam (10 ° 10.5 ’ N, 106 °. 8.9 ’ E), 0.5 m depth, 3 February 2010, collected by L.X. Tran; NSMT-P 106664, 1 specimen (male), 20.0 mm SL, collected with CTU-P 2494; NSMT-P 106665, 1 specimen (female), 22.0 mm SL, collected with CTU-P 2494; USNM 404477, 1 specimen (female), 23.8 mm SL, collected with CTU-P 2494; USNM 404478, 1 specimen (male), 20.3 mm SL, Cau Ke, Tra Vinh Province, Vietnam (9 ° 57.1 ’ N, 106 ° 1.8 ’ E), 0.5–3.5 m depths, 28 May 2010, collected by L.X. Tran; USNM 404479, 1 specimen (male, cleared and stained), 24.5 mm SL, collected with CTU-P 5020. Diagnosis. Phallostethus cuulong is distinguished from congeners in having following characters: seven serrae on the second ctenactinium in adult males (vs. five and eight in P. dunckeri and P. l e h i, respectively); 11–13 pectoral-fin rays (vs. 9–10 and 12 in P. dunckeri and P. l e h i, respectively); 11–14 + 25-26 = 37–40 vertebrae (vs. 13 + 27 = 40 and 12 + 28 = 40 in P. dunckeri and P. l e h i); approximately 5–19 teeth on paradentary (vs. 15–20 and 28 or more in P. dunckeri and P. le h i, respectively). All six examined males are dextral (vs. one and two known males of P. dunckeri are sinistral and dextral respectively, and all four know males of P. l e h i are sinistral). Description. Counts of the holotype are asterisked, and the frequency of each count is given in parentheses following the relevant count. Dorsal-fin rays 7 (2), 8 * (6) or 9 (1); anal-fin rays 24 * (4), 25 (1), 26 (3) or 27 (1); pectoral-fin rays 11 (2), 12 * (7) or 13 * (9); scales in lateral series 34 (5), 35 * (10) or 36 (3); predorsal scales 1 + 25 (1), 2 + 25 (1), 2 + 26 (4) or 2 + 27 * (3); transverse scales 6 (3), 7 * (12) or 8 * (3); circumpeduncular scales 12 * (8) or 13 (1); paradentary teeth approximately 5 (1), 6 (3), 7 (3), 8 (3), 10 (1), 13 * (1), 14 (2), 15 * (1), 18 (2) or 19 (1). The following measurements are % of SL: head length 22.1–24.1; snout length 7.0– 8.5; eye diameter 6.7–7.3; interorbital width 3.3–5.2; length of jaw 8.0– 9.4; predorsal length 78.0– 82.6; preanal length 46.4–48.7; maximum body depth 15.0– 18.7; body depth at anal-fin origin 12.4 –15.0; body width 9.4 –14.0; caudal-peduncle length 18.6–20.7; caudal-peduncle depth 5.6–7.6; length of dorsal-fin base 9.1–10.5; length of anal-fin base 32.3–36.7; pectoral-fin length 14.6–19.1; caudal-fin length 20.1–22.4. Head depressed anteriorly, with flat or barely concave interorbital space. Dorsal surface of head with membranous dome when alive or freshly collected (which can be seen in the cleared and stained specimens, e.g., Fig. 4), but shrunken and not apparent in alcohol specimen. Snout rather pointed. Eyes lateral on head, large, diameter slightly less than snout length. Mouth terminal or subterminal; jaws small, barely extend to a level of anterior margin of eye; upper jaws highly protractile. Body compressed, moderately deep. Anus and urogenital openings anterior, ventral to pectoral-fin base. A slightly frayed, fleshy membranous mid-ventral keel between urogenital opening and anal-fin origin. In males, a distinct mid-ventral groove, deepened and widened anteriorly, supports the priapium and mid-ventral keel. Pectoral fin falcate, the uppermost branched ray longest in most specimens; pectoral-fin rays branched, except for the uppermost 1 (uppermost nubby ossicle not included here; see “Materials and Methods” above) and lowermost 1–2 rays unbranched. Pelvic fin absent in males, present but rudimentary in females (Fig. 3). First dorsal fin absent; origin of second dorsal fin at, or slightly before, a level of posterior end of anal-fin base; anterior 2–3 rays simple, whereas the other rays branched. Anal fin with a long base, commencing well before midlength; anterior 2–4 rays simple, whereas the other rays branched. Caudal fin emarginate, symmetrical dorsoventrally. Male bilaterally asymmetric, dextral; namely, seminal papilla offset to right side of body (= aproctal side), and anus offset to left side of body (= proctal side); a long rod-like toxactinium curved from left to right; a large fleshy pad, the pulvinulus, covers articulation point of toxactinium and aproctal axial bone (Fig. 3). Scales on body cycloid, moderately large and deciduous; scales on abdomen largest; body entirely scaled, except for pectoral-fin base, mid-ventral groove before anal-fin origin, and mid-predorsal narrow naked space slightly behind occipital region; head and fins naked, except for posterior part of occipital region and basal part of caudal fin with some scales. Teeth on premaxilla and dentary unicuspid, slightly curved inward. Paradentary slender (as in Phallostethus dunckeri illustrated by Parenti, 1984: 4, fig. 1), with 5–19 minute teeth laterally; teeth on paradentary form a uniserial row or, in some larger specimens, biserial rows; teeth on inner row, if present, much smaller than those on outer row. Cephalic sensory canals reduced, comprising: two short infraorbital canals (each with terminal pores only) anteriorly and anteroventrally to eye; preopercular canal (with 6–7 pores). Main external bones in males including a long, curved toxactinium and a short stout ctenactinium with seven serrae dorsally (not including a hook-like distal tip); two smallest males examined (CTU-P 2493 and USNM 404478, 20.0– 20.3 mm SL) bearing 5–6 serrae on ctenactinium, assumued to represent the immature condition (and not included in the diagnosis, above). First pleural rib attached to fifth vertebra in males, fourth in females; first pleural rib in female much shorter than in male. Branchiostegal rays 4. Color when alive or freshly collected. Body subtranslucent in life, but whitish immediately after death (Fig. 1); a bright white blotch over brain when alive (assumed to fade just after death); iris silvery; minute melanophores scattered on snout, cheek and jaws; a melanophore at angle of lower jaw; a large reddish yellow blotch, slightly smaller than eye, at mid-lateral caudal fin base; male priapium with several large and small melanophores, particularly on the aproctal side (= right side in the new species) just anterior to the base of second ctenactinium; a series of minute black dots along mid-lateral septum of body musculature at least on caudal part of body; inner side of pectoral fin with many melanophores at least dorsally (the area with melanophores much more broader in females than in males); a series of mid-ventral black dots from anal-fin origin to caudal-fin base; other fins transparent. Color in alcohol. Head and body pale straw-colored; a series of irregular-sized melanophores (several of them dash-like) along midlateral septum of body musculature (at least on caudal part); paired patches of melanophores on anterior part of snout dorsally; many melanophores scattered on head above neurocranium, those posterior distinctly larger than those anterior; a melanophore at angle of lower jaw; a patch of minute gular melanophores; two patches of melanophores at throat and just behind urogenital opening in females; male priapium with several large and small melanophores, particularly on the aproctal side (= right side in this species) just anterior to the base of second ctenactinium; a mid-ventral series of black dots from anal- to caudal-fin bases, each along anal-fin base on interspace between fin rays (continuous and forming a irregular blackish gray line in some specimens); inner side of pectoral fin with many melanophores dorsally (the area with melanophores much broader in females than in males); caudal fin covered by numerous minute melanophores; other fins transparent. Distribution, habitat and the other notes. Phallostethus cuulong is known from nine specimens, six males and three females, collected from shallow waters (<1.2m depth) around banks of slow-flowing turbid canals and rivers with soft muddy bottoms in Soc Trang and Tra Vinh Provinces, Vietnam. The first author (KS) observed that a fish, latter designated as one of the paratypes of Phallostethus cuulong (NSMT-P 106665), swam slowly at the water surface around a bank of the slow-flowing tidal canal with dense semi-aquatic vegetation. A bright white blotch on dorsal surface of head was clearly confirmed in the field, but less vivid than that of the sympatric aplocheilid, Aplocheilus panchax (well-known for its reflective “pineal” spot on the top of the head). When the fish was disturbed, it quickly swam a short distance away from the original position; it was subsequently scooped up carefully using a hand net by KS. The species was usually solitary, and collected by hand nets or seine nets. Like the other phallostethids in the Vietnamese Mekong, this species has never been seen in the fish markets. As far as we aware, all fishes of the family Phallostethidae have no vernacular names in the Vietnamese Mekong (except for the new species herein named), since they are usually overlooked. Etymology. The specific name, cuulong, is the Vietnamese name of the Mekong delta (Cưu Long), where the type series of the new species was collected. The name, here applied as a noun in apposition, means “nine dragons,” in reference to nine distributaries of the Mekong basin in Vietnam. Remarks. Following the key to genera of phallostethid fishes by Parenti (1989), the new species is clearly assigned to Phallostethus by having the combination of, e.g., shield-like pulvinulus, large seminal papilla, long toxactinium, membranous dome on dorsal surface of head, 24–27 anal-fin rays, 37–40 vertebrae, serrated ctenactinium, non-projecting lower jaw beyond upper jaw, no first dorsal fin, and 7–9 second dorsal-fin rays. In particular, no other phallostethid genera are known that bear 24 or more anal-fin rays (vs. 22 or less anal-fin rays in the other phallostethids). Within the genus, the new species resembles the Bornean species Phallostethus lehi in sharing 11–13 pectoral-fin rays, but differs in having seven serrae on second ctenactinium in adult (vs. eight in P. l e h i), 25–26 caudal vertebrae (vs. 28), and 6–19 paradentary teeth (vs. 28 or more). All six examined males of the new species are dextral, immediately distinguishing them from sinistral males in P. l e h i. The new species is also distinguished from Phallostethus dunckeri, known only from Johor, Malay Peninsula but presumed to be extinct (Parenti, 1996), by having seven serrae on second ctenactinium in adults (vs. five in P. d u n c k e r i), 11–13 pectoral-fin rays (vs. 9–10), and 25–26 caudal vertebrae (vs. 27). Sexual dimorphism in the pleural ribs was reported from Phallostethus lehi by Parenti (1996); according to her, the species has the first pair of pleural ribs on the fifth vertebrae in males, the fourth vertebrae in females. This dimorphism is also found in Phallostethus cuulong. Furthermore, P. cuulong appears to show sexual dimorphism in the number of precaudal vertebrae: all six males examined have 13–14 precaudal vertebrae, as against 11–12 in all three females examined. Although Parenti & Louie (1998) reported similar sexual dimorphism in vertebral counts from four species of Neostethus, hitherto it has never been known from the other species of Phallostethus.Published as part of Shibukawa, Koichi, Tran, Dinh Dac & Tran, Loi Xuan, 2012, Phallostethus cuulong, a new species of priapiumfish (Actinopterygii: Atheriniformes: Phallostethidae) from the Vietnamese Mekong, pp. 45-51 in Zootaxa 3363 on pages 46-51, DOI: 10.5281/zenodo.28164
- …
