Your browser doesn't support javascript.
loading
Mostrar: 20 | 50 | 100
Resultados 1 - 3 de 3
Filtrar
Más filtros











Base de datos
Intervalo de año de publicación
1.
Artículo en Inglés | MEDLINE | ID: mdl-39141463

RESUMEN

Multiagent policy gradients (MAPGs), an essential branch of reinforcement learning (RL), have made great progress in both industry and academia. However, existing models do not pay attention to the inadequate training of individual policies, thus limiting the overall performance. We verify the existence of imbalanced training in multiagent tasks and formally define it as an imbalance between policies (IBPs). To address the IBP issue, we propose a dynamic policy balance (DPB) model to balance the learning of each policy by dynamically reweighting the training samples. In addition, current methods for better performance strengthen the exploration of all policies, which leads to disregarding the training differences in the team and reducing learning efficiency. To overcome this drawback, we derive a technique named weighted entropy regularization (WER), a team-level exploration with additional incentives for individuals who exceed the team. DPB and WER are evaluated in homogeneous and heterogeneous tasks, effectively alleviating the imbalanced training problem and improving exploration efficiency. Furthermore, the experimental results show that our models can outperform the state-of-the-art MAPG methods and boast over 12.1 % performance gain on average.

2.
Trials ; 24(1): 6, 2023 Jan 03.
Artículo en Inglés | MEDLINE | ID: mdl-36597103

RESUMEN

BACKGROUND: Knee osteoarthritis (OA) is a prevalent disabling disorder that involves changes in articular cartilage damage, subchondral bone remodeling, synovitis, and abnormal infrapatellar fat pad (IPFP). Due to the complicated etiology and numerous phenotypes of knee OA, limited improvement is achieved for treatments among knee OA patients with different phenotypes. Inflammatory OA phenotype is a typical knee OA phenotype, and individualized treatment targeting inflammation is a promising way to obtain an optimal therapeutic effect for people with inflammatory knee OA phenotype. Glucocorticoid is a traditional anti-inflammatory drug for knee OA, and intra-articular glucocorticoid injections are recommended clinically. However, emerging evidence has shown that repeated intra-articular glucocorticoid injections in the long term would induce cartilage loss. IPFP and its adjacent synovium are considered as the main source of inflammation in knee OA. This GLITTERS trial aims to investigate if a glucocorticoid injection into the IPFP is effective and safe over 12 weeks among knee OA patients with an inflammatory phenotype. METHODS: GLITTERS is a multicenter, double-blinded, randomized, and placebo-controlled clinical trial among knee OA patients with both Hoffa-synovitis and effusion-synovitis. Sixty participants will be allocated randomly and equally to either the glucocorticoid group or the control group. Each group will receive an injection of glucocorticoid or saline into the IPFP with an intra-articular hyaluronic acid injection as a background treatment at baseline and be followed at 4, 8, and 12 weeks. The primary outcomes will be changes in knee pain on a visual analog scale and effusion-synovitis volume measured on magnetic resonance imaging (MRI). The secondary outcomes will be changes in the total score of Western Ontario and McMaster Universities Osteoarthritis Index score, MRI-detected Hoffa-synovitis score, quality of life, pain medication use, IPFP volume, and the incidence of adverse reactions. Data analyses based on the intention-to-treat principle will include mixed-effects regressions, Wilcoxon rank-sum tests, and chi-square tests (or Fisher's exact test). DISCUSSION: GLITTERS may provide high-quality evidence for the efficacy and safety of ultrasound-guided glucocorticoid injections into IPFP among people with inflammatory knee OA in a short term. The results of this trial are expected to provide a reliable reference for a longer-term risk-benefit profile of this treatment in the future. TRIAL REGISTRATION: ClinicalTrials.gov NCT05291650. Registered on 23 March 2022.


Asunto(s)
Osteoartritis de la Rodilla , Sinovitis , Humanos , Osteoartritis de la Rodilla/terapia , Glucocorticoides/efectos adversos , Calidad de Vida , Dolor/tratamiento farmacológico , Inyecciones Intraarticulares , Sinovitis/diagnóstico por imagen , Sinovitis/tratamiento farmacológico , Sinovitis/complicaciones , Inflamación/tratamiento farmacológico , Tejido Adiposo , Resultado del Tratamiento , Ensayos Clínicos Controlados Aleatorios como Asunto , Estudios Multicéntricos como Asunto
3.
J Phys Chem Lett ; 9(11): 2751-2756, 2018 Jun 07.
Artículo en Inglés | MEDLINE | ID: mdl-29730926

RESUMEN

The interesting properties of Kagome bands, consisting of Dirac bands and a flat band, have attracted extensive attention. However, materials with only one Kagome band around the Fermi level cannot possess physical properties of Dirac Fermions and strong correlated Fermions simultaneously. Here, we propose a new type of band structure, double Kagome bands, which can realize coexistence of the two kinds of Fermions. Moreover, the new band structure is found to exist in a new two-dimensional material, phosphorus carbide P2C3. The carbide material shows good stability and unusual electronic properties. Strong magnetism appears in the structure by hole doping of the flat band, which results in spin splitting of the Dirac bands. The edge states induced by Dirac and flat bands coexist on the Fermi level, indicating outstanding transport characteristics. In addition, a possible route to experimentally grow P2C3 on some suitable substrates such as the Ag(111) surface is also discussed.

SELECCIÓN DE REFERENCIAS
DETALLE DE LA BÚSQUEDA