Neural Comput - Risk-sensitive reinforcement learning.

Tópicos

{ activ(1452) weight(1219) physic(1104) }
{ framework(1458) process(801) describ(734) }
{ take(945) account(800) differ(722) }
{ system(1050) medic(1026) inform(1018) }
{ signal(2180) analysi(812) frequenc(800) }
{ model(3404) distribut(989) bayesian(671) }
{ system(1976) rule(880) can(841) }
{ measur(2081) correl(1212) valu(896) }
{ design(1359) user(1324) use(1319) }
{ state(1844) use(1261) util(961) }
{ process(1125) use(805) approach(778) }
{ data(1737) use(1416) pattern(1282) }
{ surgeri(1148) surgic(1085) robot(1054) }
{ featur(1941) imag(1645) propos(1176) }
{ model(2341) predict(2261) use(1141) }
{ studi(1119) effect(1106) posit(819) }
{ time(1939) patient(1703) rate(768) }
{ decis(3086) make(1611) patient(1517) }
{ problem(2511) optim(1539) algorithm(950) }
{ error(1145) method(1030) estim(1020) }
{ case(1353) use(1143) diagnosi(1136) }
{ research(1085) discuss(1038) issu(1018) }
{ compound(1573) activ(1297) structur(1058) }
{ model(3480) simul(1196) paramet(876) }
{ medic(1828) order(1363) alert(1069) }
{ gene(2352) biolog(1181) express(1162) }
{ estim(2440) model(1874) function(577) }
{ can(774) often(719) complex(702) }
{ imag(1947) propos(1133) code(1026) }
{ sequenc(1873) structur(1644) protein(1328) }
{ imag(2830) propos(1344) filter(1198) }
{ network(2748) neural(1063) input(814) }
{ motion(1329) object(1292) video(1091) }
{ treatment(1704) effect(941) patient(846) }
{ learn(2355) train(1041) set(1003) }
{ studi(1410) differ(1259) use(1210) }
{ risk(3053) factor(974) diseas(938) }
{ data(2317) use(1299) case(1017) }
{ cost(1906) reduc(1198) effect(832) }
{ can(981) present(881) function(850) }
{ analysi(2126) use(1163) compon(1037) }
{ use(976) code(926) identifi(902) }
{ drug(1928) target(777) effect(648) }
{ method(1969) cluster(1462) data(1082) }
{ method(2212) result(1239) propos(1039) }
{ inform(2794) health(2639) internet(1427) }
{ imag(1057) registr(996) error(939) }
{ bind(1733) structur(1185) ligand(1036) }
{ method(1219) similar(1157) match(930) }
{ featur(3375) classif(2383) classifi(1994) }
{ imag(2675) segment(2577) method(1081) }
{ patient(2315) diseas(1263) diabet(1191) }
{ studi(2440) review(1878) systemat(933) }
{ assess(1506) score(1403) qualiti(1306) }
{ chang(1828) time(1643) increas(1301) }
{ concept(1167) ontolog(924) domain(897) }
{ clinic(1479) use(1117) guidelin(835) }
{ algorithm(1844) comput(1787) effici(935) }
{ extract(1171) text(1153) clinic(932) }
{ method(1557) propos(1049) approach(1037) }
{ data(1714) softwar(1251) tool(1186) }
{ control(1307) perform(991) simul(935) }
{ model(2220) cell(1177) simul(1124) }
{ care(1570) inform(1187) nurs(1089) }
{ general(901) number(790) one(736) }
{ method(984) reconstruct(947) comput(926) }
{ search(2224) databas(1162) retriev(909) }
{ howev(809) still(633) remain(590) }
{ data(3963) clinic(1234) research(1004) }
{ perform(999) metric(946) measur(919) }
{ import(1318) role(1303) understand(862) }
{ visual(1396) interact(850) tool(830) }
{ perform(1367) use(1326) method(1137) }
{ blood(1257) pressur(1144) flow(957) }
{ spatial(1525) area(1432) region(1030) }
{ record(1888) medic(1808) patient(1693) }
{ health(3367) inform(1360) care(1135) }
{ monitor(1329) mobil(1314) devic(1160) }
{ ehr(2073) health(1662) electron(1139) }
{ research(1218) medic(880) student(794) }
{ patient(2837) hospit(1953) medic(668) }
{ model(2656) set(1616) predict(1553) }
{ age(1611) year(1155) adult(843) }
{ group(2977) signific(1463) compar(1072) }
{ sampl(1606) size(1419) use(1276) }
{ data(3008) multipl(1320) sourc(1022) }
{ first(2504) two(1366) second(1323) }
{ intervent(3218) particip(2042) group(1664) }
{ activ(1138) subject(705) human(624) }
{ patient(1821) servic(1111) care(1106) }
{ use(2086) technolog(871) perceiv(783) }
{ health(1844) social(1437) communiti(874) }
{ structur(1116) can(940) graph(676) }
{ high(1669) rate(1365) level(1280) }
{ cancer(2502) breast(956) screen(824) }
{ use(1733) differ(960) four(931) }
{ result(1111) use(1088) new(759) }
{ implement(1333) system(1263) develop(1122) }
{ survey(1388) particip(1329) question(1065) }
{ detect(2391) sensit(1101) algorithm(908) }

Resumo

We derive a family of risk-sensitive reinforcement learning methods for agents, who face sequential decision-making tasks in uncertain environments. By applying a utility function to the temporal difference (TD) error, nonlinear transformations are effectively applied not only to the received rewards but also to the true transition probabilities of the underlying Markov decision process. When appropriate utility functions are chosen, the agents' behaviors express key features of human behavior as predicted by prospect theory (Kahneman & Tversky, 1979 ), for example, different risk preferences for gains and losses, as well as the shape of subjective probability curves. We derive a risk-sensitive Q-learning algorithm, which is necessary for modeling human behavior when transition probabilities are unknown, and prove its convergence. As a proof of principle for the applicability of the new framework, we apply it to quantify human behavior in a sequential investment task. We find that the risk-sensitive variant provides a significantly better fit to the behavioral data and that it leads to an interpretation of the subject's responses that is indeed consistent with prospect theory. The analysis of simultaneously measured fMRI signals shows a significant correlation of the risk-sensitive TD error with BOLD signal change in the ventral striatum. In addition we find a significant correlation of the risk-sensitive Q-values with neural activity in the striatum, cingulate cortex, and insula that is not present if standard Q-values are used.

Resumo Limpo

deriv famili risksensit reinforc learn method agent face sequenti decisionmak task uncertain environ appli util function tempor differ td error nonlinear transform effect appli receiv reward also true transit probabl under markov decis process appropri util function chosen agent behavior express key featur human behavior predict prospect theori kahneman tverski exampl differ risk prefer gain loss well shape subject probabl curv deriv risksensit qlearn algorithm necessari model human behavior transit probabl unknown prove converg proof principl applic new framework appli quantifi human behavior sequenti invest task find risksensit variant provid signific better fit behavior data lead interpret subject respons inde consist prospect theori analysi simultan measur fmri signal show signific correl risksensit td error bold signal chang ventral striatum addit find signific correl risksensit qvalu neural activ striatum cingul cortex insula present standard qvalu use

Resumos Similares

AMIA Annu Symp Proc - Online deviation detection for medical processes. ( 0,585400249260659 )
Int J Med Robot - The optimization of wireless power transmission: design and realization. ( 0,570377596126326 )
Comput Methods Programs Biomed - A SAS macro for target dose estimation by reinforced urn processes in phase I clinical trials. ( 0,548021599499529 )
Methods Inf Med - TADAA: Towards Automated Detection of Anaesthetic Activity. ( 0,535696881323494 )
Methods Inf Med - A business rules design framework for a pharmaceutical validation and alert system. ( 0,534401119572593 )
Int J Neural Syst - Phase synchronization of neuronal noise in mouse hippocampal epileptiform dynamics. ( 0,525057385074913 )
Comput. Biol. Med. - Hybrid fusion of linear, non-linear and spectral models for the dynamic modeling of sEMG and skeletal muscle force: an application to upper extremity amputation. ( 0,52414258490405 )
AMIA Annu Symp Proc - Studying visual behaviors from multiple eye tracking features across levels of information representation. ( 0,511866723917938 )
J Clin Monit Comput - Partitioning standard base excess: a new approach. ( 0,509938250084811 )
IEEE Trans Neural Netw Learn Syst - Fault Identification in Distributed Sensor Networks Based on Universal Probabilistic Modeling. ( 0,508826723197352 )
Med Biol Eng Comput - Adapted filter banks for feature extraction in transcranial magnetic stimulation evoked responses. ( 0,507334812802526 )
J Clin Monit Comput - Reversal of neuromuscular blockade by sugammadex does not affect EEG derived indices of depth of anesthesia. ( 0,505227311169623 )
Neural Comput - The basal ganglia optimize decision making over general perceptual hypotheses. ( 0,502320745664556 )
J Biomed Inform - Discovery of clinical pathway patterns from event logs using probabilistic topic models. ( 0,501213402789828 )
IEEE Trans Image Process - A marked point process for modeling lidar waveforms. ( 0,500772758175798 )
Int J Med Robot - Robotic interface for transabdominal division of the levators and pelvic floor reconstruction in abdominoperineal resection: a case report and technical description. ( 0,499494123469021 )
Med Biol Eng Comput - A model of tissue differentiation and bone remodelling in fractured vertebrae treated with minimally invasive percutaneous fixation. ( 0,497070601242374 )
Neural Comput - Neural mass activity, bifurcations, and epilepsy. ( 0,496691197934825 )
J Med Syst - Barriers to physical activity: a study of self-revelation in an online community. ( 0,495732994373129 )
Med Biol Eng Comput - Theoretical development and critical analysis of burst frequency equations for passive valves on centrifugal microfluidic platforms. ( 0,494767912052787 )
IEEE Trans Image Process - Wavelet variance analysis for random fields on a regular lattice. ( 0,493961487994899 )
Int J Med Inform - Faxed Arabic prescriptions: a medication error waiting to happen? ( 0,490660359792955 )
IEEE J Biomed Health Inform - RSSI/LQI-based transmission power control for body area networks in healthcare environment. ( 0,489991120054281 )
J. Med. Internet Res. - Web-based assessments of physical activity in youth: considerations for design and scale calibration. ( 0,488438959519679 )
Med Biol Eng Comput - Empirical mode decomposition and neural network for the classification of electroretinographic data. ( 0,487167285775486 )
Int J Neural Syst - Indirect adaptive control of nonlinear systems based on bilinear neuro-fuzzy approximation. ( 0,48702499586163 )
J. Med. Internet Res. - Using instructional design process to improve design and development of Internet interventions. ( 0,485463179520406 )
Med Decis Making - Estimating the relative utility of screening mammography. ( 0,484599726920356 )
J Med Syst - Privacy and security in mobile health apps: a review and recommendations. ( 0,483378206600817 )
J Clin Monit Comput - Heart rate variability indices for very short-term (30 beat) analysis. Part 2: validation. ( 0,482996941767483 )
J Med Syst - Fuzzy assessment of health information system users' security awareness. ( 0,482226603020618 )
J Biomed Inform - Incorporation of expert variability into breast cancer treatment recommendation in designing clinical protocol guided fuzzy rule system models. ( 0,481874191792946 )
IEEE Trans Vis Comput Graph - Visualizing the Variability of Gradients in Uncertain 2D Scalar Fields. ( 0,480868443939398 )
IEEE Trans Image Process - Partial differential equation-based approach for empirical mode decomposition: application on image analysis. ( 0,480239667981542 )
J Biomed Inform - Individualized patient-centered lifestyle recommendations: an expert system for communicating patient specific cardiovascular risk information and prioritizing lifestyle options. ( 0,480157765643882 )
Comput Math Methods Med - Measuring connectivity in linear multivariate processes: definitions, interpretation, and practical analysis. ( 0,479654709466455 )
J Chem Inf Model - Ranking COMMPS chemical substances by an improved POT/RLE approach. ( 0,478730845645551 )
Neural Comput - A study on L2-loss (squared hinge-loss) multiclass SVM. ( 0,478361634639247 )
Methods Inf Med - Signal interpretation from low-dosage acquisition. An investigation for computed tomographic imaging. ( 0,478086063237771 )
Comput. Biol. Med. - The impact of fine-tuning of optical recognition system on database reliability. ( 0,475611642326335 )
Comput. Biol. Med. - Statistical evaluation of coherence estimated from optimally beamformed signals. ( 0,475376714358272 )
Comput Methods Programs Biomed - ODmixed: a tool to obtain optimal designs for heterogeneous longitudinal studies with dropout. ( 0,474213100343051 )
J Am Med Inform Assoc - Protecting count queries in study design. ( 0,473066293346556 )
Comput Methods Programs Biomed - A systematic approach to embedded biomedical decision making. ( 0,472681537185046 )
Comput Methods Programs Biomed - Modeling O2 transport as an integrated system limiting (.)V(O2MAX). ( 0,472589267911306 )
Comput. Biol. Med. - Pattern discovery from patient controlled analgesia demand behavior. ( 0,472345436074385 )
J Biomed Inform - Toward automated workflow analysis and visualization in clinical environments. ( 0,471445125905627 )
J Med Syst - Hiding patients confidential datainthe ECG signal viaa transform-domain quantization scheme. ( 0,469865459615077 )
IEEE Trans Image Process - Data-free prior model for upper body pose estimation and tracking. ( 0,469551274077311 )
Neural Comput - Impact of correlated neural activity on decision-making performance. ( 0,467582937602647 )
Comput Methods Programs Biomed - Design of an efficient framework for fast prototyping of customized human-computer interfaces and virtual environments for rehabilitation. ( 0,466389275687492 )
Med Decis Making - A comparison of direct and indirect methods for the estimation of health utilities from clinical outcomes. ( 0,466153571739572 )
Lifetime Data Anal - Estimation of finite population duration distributions from longitudinal survey panels with intermittent followup. ( 0,464927451353937 )
Neural Comput - Identification of directed influence: Granger causality, Kullback-Leibler divergence, and complexity. ( 0,464078023342098 )
Comput Methods Programs Biomed - Comparison of checking behavior in adults with or without checking symptom of obsessive-compulsive disorder using a novel computer-based measure. ( 0,463878161511819 )
Comput Math Methods Med - A general framework for modeling sub- and ultraharmonics of ultrasound contrast agent signals with MISO volterra series. ( 0,463747464148398 )
Comput Math Methods Med - Disease control through voluntary vaccination decisions based on the smoothed best response. ( 0,463295619734112 )
J Biomed Inform - Intelligent mobile support for therapy adherence and behavior change. ( 0,462078352097895 )
Lifetime Data Anal - Quadratic fractional age assumption revisited. ( 0,460923139565803 )
Artif Intell Med - How to use contextual knowledge in medical case-based reasoning systems: a survey on very recent trends. ( 0,460626557308071 )
J. Comput. Biol. - Probabilistic cellular automata. ( 0,460544136624478 )
Lifetime Data Anal - The versatility of multi-state models for the analysis of longitudinal data with unobservable features. ( 0,460512605569328 )
J. Med. Internet Res. - Comparison of physical activity measures using mobile phone-based CalFit and Actigraph. ( 0,459584246577359 )
Comput Methods Programs Biomed - A comparative study between a simplified Kalman filter and Sliding Window Averaging for single trial dynamical estimation of event-related potentials. ( 0,45882949640488 )
Comput Math Methods Med - Cross time-frequency analysis for combining information of several sources: application to estimation of spontaneous respiratory rate from photoplethysmography. ( 0,458485676201014 )
Comput Methods Programs Biomed - PopED: an extended, parallelized, nonlinear mixed effects models optimal design tool. ( 0,458476468580457 )
Comput. Biol. Med. - New approach to predicting proconvulsant activity with the use of Support Vector Regression. ( 0,45804494490778 )
Neural Comput - Online variational inference for state-space models with point-process observations. ( 0,456932479787074 )
J Clin Monit Comput - Bi-spectral index, entropy and predicted plasma propofol concentrations with target controlled infusions in Indian patients. ( 0,45682445352973 )
Med Decis Making - Management of undescended testis: a decision analysis. ( 0,456217356871199 )
AMIA Annu Symp Proc - A healthcare utilization analysis framework for hot spotting and contextual anomaly detection. ( 0,454794005657451 )
Int J Med Inform - Designing personal exercise monitoring employing multiple modes of delivery: implications from a qualitative study on heart rate monitoring. ( 0,453876770691516 )
J Biomed Inform - Exploring medical device design and use through layers of distributed cognition: how a glucometer is coupled with its context. ( 0,453585730731681 )
J. Med. Internet Res. - A conceptual framework and principles for trusted pervasive health. ( 0,452923080861063 )
Med Decis Making - Protocols for time tradeoff valuations of health states worse than dead: a literature review. ( 0,451712108294804 )
Comput. Biol. Med. - Improved explanation of human intelligence using cortical features with second order moments and regression. ( 0,451657146147583 )
IEEE Trans Image Process - Nonlocal PDEs-based morphology on weighted graphs for image and data processing. ( 0,451349781939096 )
Int J Comput Assist Radiol Surg - MITK-US: real-time ultrasound support within MITK. ( 0,451204821433369 )
AMIA Annu Symp Proc - An Empirical Validation of Recursive Noisy OR (RNOR) Rule for Asthma Prediction. ( 0,45119331892414 )
Med Biol Eng Comput - Remote sensing of patterns of cardiac activity on an ambulatory subject using millimeter-wave interferometry and statistical methods. ( 0,450421051285306 )
Med Biol Eng Comput - A muscle architecture model offering control over motor unit fiber density distributions. ( 0,45019840411048 )
J Med Syst - Design and development of a heart rate variability analyzer. ( 0,450183212908958 )
Lifetime Data Anal - Assessing discrimination of risk prediction rules in a clustered data setting. ( 0,449988568280122 )
BMC Med Inform Decis Mak - Generation and application of drug indication inference models using typed network motif comparison analysis. ( 0,449260703414845 )
Wiley Interdiscip Rev Syst Biol Med - Simulation of cellular biochemical system kinetics. ( 0,448798536437267 )
IEEE J Biomed Health Inform - Optimization of heartbeat detection in fiber-optic unobtrusive measurements by using maximum a posteriori probability estimation. ( 0,447854223117332 )
Res Synth Methods - Meta-analysis and the reversed Theorem of the Means. ( 0,447789557537099 )
J Chem Inf Model - Conformation-dependent QSPR models: logPOW. ( 0,447582935056762 )
Int J Neural Syst - Analysis of absence seizure generation using EEG spatial-temporal regularity measures. ( 0,446747587472743 )
Comput. Biol. Med. - EEG-based emotion estimation using Bayesian weighted-log-posterior function and perceptron convergence algorithm. ( 0,446628003456133 )
IEEE Trans Image Process - Image reduction using means on discrete product lattices. ( 0,445495890170328 )
J. Med. Internet Res. - Measures of physical activity using cell phones: validation using criterion methods. ( 0,445340449487096 )
Comput Math Methods Med - Robust joint analysis with data fusion in two-stage quantitative trait genome-wide association studies. ( 0,444475299988615 )
Comput Methods Programs Biomed - Individualized model discovery: the case of anemia patients. ( 0,444444842686266 )
J Am Med Inform Assoc - Patient-initiated electronic health record amendment requests. ( 0,443523497840073 )
IEEE Trans Image Process - A geometric method for optimal design of color filter arrays. ( 0,443511528209472 )
Comput Methods Programs Biomed - QRS detection using S-Transform and Shannon energy. ( 0,443341069520925 )
IEEE Trans Image Process - A Kullback-Leibler divergence approach to blind image restoration. ( 0,443269301523658 )
IEEE Trans Neural Netw Learn Syst - GrDHP: a general utility function representation for dual heuristic dynamic programming. ( 0,442916647225676 )
Int J Med Inform - Development and alarm threshold evaluation of a side rail integrated sensor technology for the prevention of falls. ( 0,441955805924788 )