251 results on '"Gangwani P"'
Search Results
2. Aspects of the Higgs phase in SU(2)xU(1) lattice gauge Higgs theory
- Author
-
Gangwani, Shivam, Greensite, Jeff, and Oualla, Anass
- Subjects
High Energy Physics - Lattice - Abstract
Using a simplified lattice version of the electroweak sector of the standard model, with dynamical fermions excluded, we determine at fixed Weinberg angle the transition line between the confined phase and the Higgs phase, the latter defined as the region where the global center subgroup of the gauge group is spontaneously broken, and "separation of charge" confinement disappears. We then search, via lattice Monte Carlo simulations, for possible neutral vector bosons in the Higgs region, apart from the photon and Z. There are numerical indications of a "light Z" in the lattice data (along with the photon and the Z), but a lack of the expected scaling of the light mass particle excludes any firm conclusions about the physical spectrum., Comment: 10 pages, 11 figures. v2: expanded citation to prior work
- Published
- 2023
3. A Comprehensive Analysis of Reported Adverse Events and Device Failures Associated with Esophageal Self-Expandable Metal Stents: An FDA MAUDE Database Study
- Author
-
Jaber, Fouad, Alsakarneh, Saqr, Alsharaeh, Tala, Salahat, Ahmed-Jordan, Jaber, Mohammad, Mohamed, Islam, Gangwani, Manesh Kumar, Aldiabat, Mohammad, Kilani, Yassine, Ahmed, Mohamed, Madi, Mahmoud, Numan, Laith, and Bazarbashi, Ahmad Najdat
- Published
- 2024
- Full Text
- View/download PDF
4. Sarcopenia is a risk factor for post-transjugular intrahepatic portosystemic shunt hepatic encephalopathy and mortality: A systematic review and meta-analysis
- Author
-
Ahmed, Zohaib, Badal, Joyce, Gangwani, Manesh Kumar, Nawaz, Ahmad, Badal, Bryan, Arif, Syeda Faiza, Farooq, Umer, Kamal, Faisal, Javaid, Toseef, Aziz, Muhammad, Lee-Smith, Wade, Mahmood, Asif, Merza, Nooraldin, Kobeissy, Abdallah, Nawras, Ali, and Hassan, Mona
- Published
- 2024
- Full Text
- View/download PDF
5. Multi-Objective Optimization via Wasserstein-Fisher-Rao Gradient Flow
- Author
-
Ren, Yinuo, Xiao, Tesi, Gangwani, Tanmay, Rangi, Anshuka, Rahmanian, Holakou, Ying, Lexing, and Sanyal, Subhajit
- Subjects
Computer Science - Machine Learning ,Mathematics - Optimization and Control ,Statistics - Machine Learning - Abstract
Multi-objective optimization (MOO) aims to optimize multiple, possibly conflicting objectives with widespread applications. We introduce a novel interacting particle method for MOO inspired by molecular dynamics simulations. Our approach combines overdamped Langevin and birth-death dynamics, incorporating a "dominance potential" to steer particles toward global Pareto optimality. In contrast to previous methods, our method is able to relocate dominated particles, making it particularly adept at managing Pareto fronts of complicated geometries. Our method is also theoretically grounded as a Wasserstein-Fisher-Rao gradient flow with convergence guarantees. Extensive experiments confirm that our approach outperforms state-of-the-art methods on challenging synthetic and real-world datasets.
- Published
- 2023
6. ppHiC: Interactive exploration of Hi-C results on the ProteinPaint web portal
- Author
-
Akanksha Rajput, Colleen Reilly, Airen Zaldivar Peraza, Jian Wang, Edgar Sioson, Gavriel Matt, Robin Paul, Congyu Lu, Aleksandar Acic, Karishma Gangwani, and Xin Zhou
- Subjects
Hi-C ,ProteinPaint ,Contact matrix ,Visualization ,Web server ,Genomic rearrangement ,Biotechnology ,TP248.13-248.65 - Abstract
The ProteinPaint Hi-C tool (ppHiC) facilitates web-based visualization and collaborative exploration of Hi-C data, a vital resource for understanding three-dimensional genomic structures. ppHiC allows researchers to easily analyze large Hi-C datasets on a web browser without requiring the computational expertise that has heretofore limited access to this complex genomic data. The platform is compatible with multiple Hi-C data versions and boasts a highly customizable interface, including a configuration panel for the precise adjustment of key visualization parameters. The tool’s interactive features offer a broad range of views, from whole-genome landscapes to detailed interactions between pairs of loci, that are accessible within a single, integrated environment. Here, we demonstrate how using ppHiC to visualize an altered chromatin conformational landscape in neuroblastoma can inform understanding of the genomic rearrangements in this cancer.
- Published
- 2024
- Full Text
- View/download PDF
7. Single Versus Second Observer vs Artificial Intelligence to Increase the ADENOMA Detection Rate of Colonoscopy—A Network Analysis
- Author
-
Gangwani, Manesh Kumar, Haghbin, Hossein, Ishtiaq, Rizwan, Hasan, Fariha, Dillard, Julia, Jaber, Fouad, Dahiya, Dushyant Singh, Ali, Hassam, Salim, Shaharyar, Lee-Smith, Wade, Sohail, Amir Humza, Inamdar, Sumant, Aziz, Muhammad, and Hart, Benjamin
- Published
- 2024
- Full Text
- View/download PDF
8. Can my surgeon scope? Trends in endoscopy training volume and experience among general surgery residents in the United States: a nationwide analysis
- Author
-
Sohail, Amir Humza, Martinez, Christian, Martinez, Kevin, Nguyen, Hoang, Flesner, Samuel, Khan, Abdullah, Quazi, Mohammed A., Rasheed, Waqas, Ali, Hassam, Dahiya, Dushyant Singh, Gangwani, Manesh Kumar, Aziz, Muhammad, and Goyal, Aman
- Published
- 2024
- Full Text
- View/download PDF
9. Selective Uncertainty Propagation in Offline RL
- Author
-
Krishnamurthy, Sanath Kumar, Modi, Shrey, Gangwani, Tanmay, Katariya, Sumeet, Kveton, Branislav, and Rangi, Anshuka
- Subjects
Computer Science - Machine Learning ,Computer Science - Artificial Intelligence - Abstract
We consider the finite-horizon offline reinforcement learning (RL) setting, and are motivated by the challenge of learning the policy at any step h in dynamic programming (DP) algorithms. To learn this, it is sufficient to evaluate the treatment effect of deviating from the behavioral policy at step h after having optimized the policy for all future steps. Since the policy at any step can affect next-state distributions, the related distributional shift challenges can make this problem far more statistically hard than estimating such treatment effects in the stochastic contextual bandit setting. However, the hardness of many real-world RL instances lies between the two regimes. We develop a flexible and general method called selective uncertainty propagation for confidence interval construction that adapts to the hardness of the associated distribution shift challenges. We show benefits of our approach on toy environments and demonstrate the benefits of these techniques for offline policy learning.
- Published
- 2023
10. Is serum brain-derived neurotrophic factor related to craving for or use of alcohol, cocaine, or methamphetamine?
- Author
-
Gangwani P, Patel G, Singh M, Underwood WA, Nejtek VA, Hilburn C, and Forster MJ
- Subjects
Neurosciences. Biological psychiatry. Neuropsychiatry ,RC321-571 ,Neurology. Diseases of the nervous system ,RC346-429 - Abstract
Craig Hilburn, Vicki A Nejtek, Wendy A Underwood, Meharvan Singh, Gauravkumar Patel, Pooja Gangwani, Michael J ForsterUniversity of North Texas Health Science Center at Fort Worth, TX, USABackground: Data suggests that brain-derived neurotropic factor (BDNF) plays a neuroadaptive role in addiction. Whether serum BDNF levels are different in alcohol or psychostimulants as a function of craving is unknown. Here, we examined craving and serum BDNF levels in persons with alcohol versus psychostimulant dependence. Our goals were to explore BDNF as an objective biomarker for 1) craving 2) abstinence, and 3) years of chronic substance use.Methods: An exploratory, cross-sectional study was designed. Men and women between 20–65 years old with alcohol, cocaine, or methamphetamine dependence were eligible. A craving questionnaire was used to measure alcohol, cocaine and methamphetamine cravings. Serum levels of BDNF were measured using enzyme linked immunoassay. Analysis of variance, chi-square, and correlations were performed using a 95% confidence interval and a significance level of P < 0.05.Results: We found a significant difference in the mean craving score among alcohol, cocaine and methamphetamine dependent subjects. There were no significant influences of race, gender, psychiatric disorder or psychotropic medication on serum BDNF levels. We found that among psychostimulant users BDNF levels were significantly higher in men than in women when the number of abstinent days was statistically controlled. Further, a significant correlation between serum BDNF levels and the number of abstinent days since last psychostimulant use was found.Conclusion: These data suggest that BDNF may be a biomarker of abstinence in psychostimulant dependent subjects and inform clinicians about treatment initiatives. The results are interpreted with caution due to small sample size and lack of a control group.Keywords: BDNF, alcohol, cocaine, methamphetamine, craving
- Published
- 2011
11. Relationships of hospitalization outcomes and timing to endoscopy in non-variceal upper gastrointestinal bleeding: A nationwide analysis
- Author
-
Weissman, Simcha, Aziz, Muhammad, Bangolo, Ayrton I, Ehrlich, Dean, Forlemu, Arnold, Willie, Anthony, Gangwani, Manesh K, Waqar, Danish, Terefe, Hannah, Singh, Amritpal, Gonzalez, Diego MC, Sajja, Jayadev, Emiroglu, Fatma L, Dinko, Nicholas, Mohamed, Ahmed, Fallorina, Mark A, Kosoy, David, Shenoy, Ankita, Nanavati, Anvit, Feuerstein, Joseph D, and Tabibian, James H
- Subjects
Patient Safety ,Health Services ,Clinical Research ,Good Health and Well Being ,Upper gastrointestinal bleeding ,Esophagogastroduodenoscopy ,Outcomes ,Mortality ,Anticoagulation - Abstract
BackgroundThe optimal timing of esophagogastroduodenoscopy (EGD) and the impact of clinico-demographic factors on hospitalization outcomes in non-variceal upper gastrointestinal bleeding (NVUGIB) remains an area of active research.AimTo identify independent predictors of outcomes in patients with NVUGIB, with a particular focus on EGD timing, anticoagulation (AC) status, and demographic features.MethodsA retrospective analysis of adult patients with NVUGIB from 2009 to 2014 was performed using validated ICD-9 codes from the National Inpatient Sample database. Patients were stratified by EGD timing relative to hospital admission (≤ 24 h, 24-48 h, 48-72 h, and > 72 h) and then by AC status (yes/no). The primary outcome was all-cause inpatient mortality. Secondary outcomes included healthcare usage.ResultsOf the 1082516 patients admitted for NVUGIB, 553186 (51.1%) underwent EGD. The mean time to EGD was 52.8 h. Early (< 24 h from admission) EGD was associated with significantly decreased mortality, less frequent intensive care unit admission, shorter length of hospital stays, lower hospital costs, and an increased likelihood of discharge to home (all with P < 0.001). AC status was not associated with mortality among patients who underwent early EGD (aOR 0.88, P = 0.193). Male sex (OR 1.30) and Hispanic (OR 1.10) or Asian (aOR 1.38) race were also independent predictors of adverse hospitalization outcomes in NVUGIB.ConclusionBased on this large, nationwide study, early EGD in NVUGIB is associated with lower mortality and decreased healthcare usage, irrespective of AC status. These findings may help guide clinical management and would benefit from prospective validation.
- Published
- 2023
12. Design Optimization of Water Distribution Networks with Dynamic Search Space Reduction GA
- Author
-
Gangwani, Laxmi, Dongre, Shilpa, Gupta, Rajesh, Abdy Sayyed, Mohd Abbas H., and Tanyimboh, Tiku
- Published
- 2024
- Full Text
- View/download PDF
13. The impact of COVID-19 on hospitalizations that underwent endoscopic retrograde cholangiopancreatography in the United States
- Author
-
Dahiya, Dushyant Singh, Pinnam, Bhanu Siva Mohan, Chandan, Saurabh, Gangwani, Manesh Kumar, Ali, Hassam, Deliwala, Smit, Bapaye, Jay, Aziz, Muhammad, Merza, Nooraldin, Inamdar, Sumant, Al-Haddad, Mohammad, and Sharma, Neil
- Published
- 2024
- Full Text
- View/download PDF
14. Perception and practice of the eight limbs of yoga in yoga teachers: A cross-sectional descriptive study
- Author
-
Kaushal Kumar Alam, Nonita Gangwani, and Mamta Mohan
- Subjects
astanga yoga ,eight limbs of yoga ,yoga ,yoga teachers ,Medicine - Abstract
Background: Yoga is an ancient wisdom that originated in India and encompasses eight limbs, including yama, niyama, asanas, pranayama, pratyahara, dharana, dhyana, and samadhi. Despite its holistic nature, many studies on yoga tend to focus solely on its physical aspects, breathing practices, and meditation, neglecting other essential components. To address this gap, a study was conducted to determine the perception and practice of all eight limbs of yoga among yoga teachers. Method: A self-designed questionnaire containing 28 items was used for assessing the knowledge, perception, and practice of the eight limbs of yoga among yoga teachers. A total of 37 yoga teachers participated in the study. Cross-tabulation was performed to analyze the data. Results: The majority of participants (>80%) in this study on the practice of astanga yoga reported that all limbs of yoga were essential in their practice. However, there were varying degrees of emphasis placed on each limb. Pranayama (91.9%) and asana (89.2%) were given the most emphasis, while niyama (75.7%), yama (73%), pratyahara (70.3%), dhyana (70.3%), and dharana (64.9%) were given slightly less emphasis. Interestingly, those with more experience placed greater emphasis on various aspects of yoga. Conclusion: Practitioners should embrace all eight limbs of yoga when introducing it to new populations to increase access to yoga. This will help promote the benefits of yoga and make it more accessible to those who may benefit from it. Health professionals should have a comprehensive understanding of the holistic practice of yoga, including its ethics, postures, breath, mindfulness, and meditation, to promote its maximum benefits and avoid potentially harmful practices.
- Published
- 2024
- Full Text
- View/download PDF
15. Neuronal and astrocytic contributions to Huntington’s disease dissected with zinc finger protein transcriptional repressors
- Author
-
Gangwani, Mohitkumar R, Soto, Joselyn S, Jami-Alahmadi, Yasaman, Tiwari, Srushti, Kawaguchi, Riki, Wohlschlegel, James A, and Khakh, Baljit S
- Subjects
Biochemistry and Cell Biology ,Genetics ,Biological Sciences ,Neurosciences ,Biotechnology ,Huntington's Disease ,Brain Disorders ,Neurodegenerative ,Rare Diseases ,Underpinning research ,1.1 Normal biological development and functioning ,2.1 Biological and endogenous factors ,Aetiology ,Neurological ,Animals ,Huntington Disease ,Astrocytes ,Huntingtin Protein ,Neurons ,Transcription Factors ,Zinc Fingers ,Mutant Proteins ,Disease Models ,Animal ,CP: Neuroscience ,Huntington’s disease ,astrocyte ,huntingtin ,medium spiny neuron ,neurodegeneration ,striatum ,therapeutics ,zinc finger protein ,Medical Physiology ,Biological sciences - Abstract
Huntington's disease (HD) is caused by expanded CAG repeats in the huntingtin gene (HTT) resulting in expression of mutant HTT proteins (mHTT) with extended polyglutamine tracts, including in striatal neurons and astrocytes. It is unknown whether pathophysiology in vivo can be attenuated by lowering mHTT in either cell type throughout the brain, and the relative contributions of neurons and astrocytes to HD remain undefined. We use zinc finger protein (ZFP) transcriptional repressors to cell-selectively lower mHTT in vivo. Astrocytes display loss of essential functions such as cholesterol metabolism that are partly driven by greater neuronal dysfunctions, which encompass neuromodulation, synaptic, and intracellular signaling pathways. Using transcriptomics, proteomics, electrophysiology, and behavior, we dissect neuronal and astrocytic contributions to HD pathophysiology. Remarkably, brain-wide delivery of neuronal ZFPs results in strong mHTT lowering, rescue of HD-associated behavioral and molecular phenotypes, and significant extension of lifespan, findings that support translational development.
- Published
- 2023
16. Imitation Learning from Observations under Transition Model Disparity
- Author
-
Gangwani, Tanmay, Zhou, Yuan, and Peng, Jian
- Subjects
Computer Science - Machine Learning ,Statistics - Machine Learning - Abstract
Learning to perform tasks by leveraging a dataset of expert observations, also known as imitation learning from observations (ILO), is an important paradigm for learning skills without access to the expert reward function or the expert actions. We consider ILO in the setting where the expert and the learner agents operate in different environments, with the source of the discrepancy being the transition dynamics model. Recent methods for scalable ILO utilize adversarial learning to match the state-transition distributions of the expert and the learner, an approach that becomes challenging when the dynamics are dissimilar. In this work, we propose an algorithm that trains an intermediary policy in the learner environment and uses it as a surrogate expert for the learner. The intermediary policy is learned such that the state transitions generated by it are close to the state transitions in the expert dataset. To derive a practical and scalable algorithm, we employ concepts from prior work on estimating the support of a probability distribution. Experiments using MuJoCo locomotion tasks highlight that our method compares favorably to the baselines for ILO with transition dynamics mismatch., Comment: ICLR 2022 camera-ready
- Published
- 2022
17. Correction: Single Versus Second Observer vs Artificial Intelligence to Increase the ADENOMA Detection Rate of Colonoscopy—A Network Analysis
- Author
-
Gangwani, Manesh Kumar, Haghbin, Hossein, Ishtiaq, Rizwan, Hasan, Fariha, Dillard, Julia, Jaber, Fouad, Dahiya, Dushyant Singh, Ali, Hassam, Salim, Shaharyar, Lee‑Smith, Wade, Sohail, Amir Humza, Inamdar, Sumant, Aziz, Muhammad, and Hart, Benjamin
- Published
- 2024
- Full Text
- View/download PDF
18. MicroRNA-502-3p regulates GABAergic synapse function in hippocampal neurons
- Author
-
Bhupender Sharma, Melissa M Torres, Sheryl Rodriguez, Laxman Gangwani, and Subodh Kumar
- Subjects
alzheimer’s disease ,gabaergic synapse ,gamma-aminobutyric acid type a receptor subunit α-1 (gabrα1) ,microrna-502-3p (mir-502-3p) ,mirna in situ hybridization ,patch-clamp ,Neurology. Diseases of the nervous system ,RC346-429 - Abstract
Gamma-aminobutyric acid (GABA)ergic neurons, the most abundant inhibitory neurons in the human brain, have been found to be reduced in many neurological disorders, including Alzheimer’s disease and Alzheimer’s disease-related dementia. Our previous study identified the upregulation of microRNA-502-3p (miR-502-3p) and downregulation of GABA type A receptor subunit α-1 in Alzheimer’s disease synapses. This study investigated a new molecular relationship between miR-502-3p and GABAergic synapse function. In vitro studies were performed using the mouse hippocampal neuronal cell line HT22 and miR-502-3p agomiRs and antagomiRs. In silico analysis identified multiple binding sites of miR-502-3p at GABA type A receptor subunit α-1 mRNA. Luciferase assay confirmed that miR-502-3p targets the GABA type A receptor subunit α-1 gene and suppresses the luciferase activity. Furthermore, quantitative reverse transcription-polymerase chain reaction, miRNA in situ hybridization, immunoblotting, and immunostaining analysis confirmed that overexpression of miR-502-3p reduced the GABA type A receptor subunit α-1 level, while suppression of miR-502-3p increased the level of GABA type A receptor subunit α-1 protein. Notably, as a result of the overexpression of miR-502-3p, cell viability was found to be reduced, and the population of necrotic cells was found to be increased. The whole cell patch-clamp analysis of human-GABA receptor A-α1/β3/γ2L human embryonic kidney (HEK) recombinant cell line also showed that overexpression of miR-502-3p reduced the GABA current and overall GABA function, suggesting a negative correlation between miR-502-3p levels and GABAergic synapse function. Additionally, the levels of proteins associated with Alzheimer’s disease were high with miR-502-3p overexpression and reduced with miR-502-3p suppression. The present study provides insight into the molecular mechanism of regulation of GABAergic synapses by miR-502-3p. We propose that micro-RNA, in particular miR-502-3p, could be a potential therapeutic target to modulate GABAergic synapse function in neurological disorders, including Alzheimer’s disease and Alzheimer’s disease-related dementia.
- Published
- 2024
- Full Text
- View/download PDF
19. Hindsight Foresight Relabeling for Meta-Reinforcement Learning
- Author
-
Wan, Michael, Peng, Jian, and Gangwani, Tanmay
- Subjects
Computer Science - Machine Learning ,Statistics - Machine Learning - Abstract
Meta-reinforcement learning (meta-RL) algorithms allow for agents to learn new behaviors from small amounts of experience, mitigating the sample inefficiency problem in RL. However, while meta-RL agents can adapt quickly to new tasks at test time after experiencing only a few trajectories, the meta-training process is still sample-inefficient. Prior works have found that in the multi-task RL setting, relabeling past transitions and thus sharing experience among tasks can improve sample efficiency and asymptotic performance. We apply this idea to the meta-RL setting and devise a new relabeling method called Hindsight Foresight Relabeling (HFR). We construct a relabeling distribution using the combination of "hindsight", which is used to relabel trajectories using reward functions from the training task distribution, and "foresight", which takes the relabeled trajectories and computes the utility of each trajectory for each task. HFR is easy to implement and readily compatible with existing meta-RL algorithms. We find that HFR improves performance when compared to other relabeling methods on a variety of meta-RL tasks., Comment: ICLR 2022 camera-ready
- Published
- 2021
20. Modeling and prediction of business success: a survey
- Author
-
Gangwani, Divya and Zhu, Xingquan
- Published
- 2024
- Full Text
- View/download PDF
21. The overall quality of evidence of recommendations surrounding nutrition and diet in inflammatory bowel disease
- Author
-
Weissman, Simcha, Fung, Brian M., Bangolo, Ayrton, Rashid, Atif, Khan, Badar F., Tirumala, Aditya K. Gudimella, Nagpaul, Sneha, Cornwell, Samuel, Karamthoti, Praveena, Murugan, Vignesh, Taranichi, Ihsan S., Kalinin, Maksim, Wishart, Annetta, Khalaf, Ibtihal, Kodali, Naga A., Aluri, Pruthvi S. C., Kejela, Yabets, Abdul, Rub, Jacob, Feba M., Manoharasetty, Advaith, Sethi, Aparna, Nadimpallli, Preethi M., Ballestas, Natalia P., Venkatraman, Aarushi, Chirumamilla, Avinash, Nagesh, Vignesh K., Gangwani, Manesh K., Issokson, Kelly, Aziz, Muhammad, Swaminath, Arun, and Feuerstein, Joseph D.
- Published
- 2023
- Full Text
- View/download PDF
22. Exploring investor-business-market interplay for business success prediction
- Author
-
Gangwani, Divya, Zhu, Xingquan, and Furht, Borko
- Published
- 2023
- Full Text
- View/download PDF
23. Adaptive Verifiable Coded Computing: Towards Fast, Secure and Private Distributed Machine Learning
- Author
-
Tang, Tingting, Ali, Ramy E., Hashemi, Hanieh, Gangwani, Tynan, Avestimehr, Salman, and Annavaram, Murali
- Subjects
Computer Science - Distributed, Parallel, and Cluster Computing ,Computer Science - Cryptography and Security ,Computer Science - Information Theory ,Computer Science - Machine Learning - Abstract
Stragglers, Byzantine workers, and data privacy are the main bottlenecks in distributed cloud computing. Some prior works proposed coded computing strategies to jointly address all three challenges. They require either a large number of workers, a significant communication cost or a significant computational complexity to tolerate Byzantine workers. Much of the overhead in prior schemes comes from the fact that they tightly couple coding for all three problems into a single framework. In this paper, we propose Adaptive Verifiable Coded Computing (AVCC) framework that decouples the Byzantine node detection challenge from the straggler tolerance. AVCC leverages coded computing just for handling stragglers and privacy, and then uses an orthogonal approach that leverages verifiable computing to mitigate Byzantine workers. Furthermore, AVCC dynamically adapts its coding scheme to trade-off straggler tolerance with Byzantine protection. We evaluate AVCC on a compute-intensive distributed logistic regression application. Our experiments show that AVCC achieves up to $4.2\times$ speedup and up to $5.1\%$ accuracy improvement over the state-of-the-art Lagrange coded computing approach (LCC). AVCC also speeds up the conventional uncoded implementation of distributed logistic regression by up to $7.6\times$, and improves the test accuracy by up to $12.1\%$.
- Published
- 2021
- Full Text
- View/download PDF
24. Trigger Films to Teach Core Competencies of Ethics and Professionalism to First-Year Medical and Nursing Students
- Author
-
Gangwani, Nonita, Singh, Satendra, and Khaliq, Farah
- Abstract
Professionalism and communication were formally introduced in India's Competency-Based Curriculum (CBC) as part of the five roles of an Indian medical graduate and 10 core competencies in the Bachelor of Science Nursing program. It may be challenging to teach the complexity of clinical medical ethics to undergraduate students at a young age in the absence of direct patient interaction. Trigger films (TFs) are brief (3-10 min) clips that have been used in the West to provoke debate, promote reflection, and assist trainees in dealing with ethical dilemmas. The aim of this study was to determine whether TFs can be used to teach professionalism and ethics to undergraduate medical and nursing students as an innovative and interesting tool and to see whether this results in any changes in knowledge. A 2-h module supported by an introductory PowerPoint presentation and using four TFs on the four pillars of ethics (beneficence, nonmaleficence, autonomy, and justice) was developed and piloted in the foundation course for the new cohort of medical and nursing students. Quantitative, open-ended feedback was taken from learners after module delivery, and knowledge was assessed with a retrospective pre-post questionnaire. The majority of students found TFs an innovative and interesting tool to teach medical ethics. There was a gain in knowledge of autonomy (52%), beneficence (48%), nonmaleficence (46%), and justice (38%). TFs can be effective tools to impart core competencies in ethics and professionalism to both nursing and medical students in the new CBC.
- Published
- 2022
- Full Text
- View/download PDF
25. Interstitial Ectopic Pregnancy with Enhanced Myometrial Vascularity: A Rare Case Successfully Treated with Uterine Artery Embolization
- Author
-
Navpreet Kaur Khurana, Gaurav Gangwani, Kamalapriya Thiyagarajan, Aditi Chaurasia, and Akanksha Thakre
- Subjects
uterine artery embolization ,enhanced myometrial vascularity ,ectopic pregnancy ,Medical physics. Medical radiology. Nuclear medicine ,R895-920 - Abstract
Enhanced myometrial vascularity (EMV) is a rare disorder associated with various obstetrical and gynecological pathologies. We describe a unique case of interstitial ectopic pregnancy associated with EMV successfully managed with bilateral uterine artery embolization.
- Published
- 2023
- Full Text
- View/download PDF
26. Impact of Religious Beliefs and Religious Practices on Individuals' Ability to Cope up With Covid-19 Pandemic: A Study with Special Reference to Followers of Islam in Saudi Arabia
- Author
-
Sanjeevni Gangwani, Shaimaa Ballout, and Nourah Alhasawi
- Subjects
Islam. Bahai Faith. Theosophy, etc. ,BP1-610 - Abstract
The current study examined the impact of religious beliefs and practices implemented by Muslims in Saudi Arabia on their ability to cope with stress, social isolation, uncertainty, and providing social support during the COVID-19 pandemic. The study included a questionnaire with a sample of 1,511 respondents. The relationship of demographic variables and religious beliefs and practices of respondents was analyzed using a correlation table. It was found that gender and civil status positively correlated with religious beliefs and practices. Whereas nationality, location, employment status, and monthly income had no significant impact on religious beliefs and practices. The study revealed a positive association of religious beliefs and practices on an individual's ability to cope with stress, social isolation, and uncertainty during the COVID-19 pandemic. Some interventions, such as using social media, helped them cope as well.
- Published
- 2023
- Full Text
- View/download PDF
27. Comparable Efficacy for Push Versus Pull Technique in Esophageal Food Impaction: Systematic Review with Meta-Analysis
- Author
-
Gangwani, Manesh Kumar, Aziz, Muhammad, Dahiya, Dushyant Singh, Aziz, Abeer, Priyanka, Fnu, Karna, Rahul, Lee-Smith, Wade, Ahmed, Zohaib, Kamal, Faisal, Inamdar, Sumant, Alastal, Yaseen, and Adler, Douglas
- Published
- 2023
- Full Text
- View/download PDF
28. Documenting the Recovery of Vascular Services in European Centres Following the Initial COVID-19 Pandemic Peak: Results from a Multicentre Collaborative Study
- Author
-
collaborative, The VERN COVER study, Ruffino, Maria Antonella, Chan, Sharon, Coughlin, Patrick, Awopetu, Ayoola, Stather, Philip, Lane, Tristan, Theodosiou, Dimitrios, Ahmed, Mohamed Abozeid, Vasudevan, Thodur, Ibrahim, Mohammed, Maadany, Faraj Al, Eljareh, Mohamed, Alkhafeefi, Fatimah Saad, Coscas, Raphael, Ünal, Ertekin Utku, Pulli, Raffaele, Zacà, Sergio, Angiletta, Domenico, Kotsis, Thomas, Moawad, Magdy, Tozzi, Matteo, Patelis, Nikolaos, Lazaris, Andreas M, Chuen, Jason, Croo, Alexander, Tsolaki, Elpiniki, Zenunaj, Gladiol, Kamal, Dhafer, Tolba, Mahmoud MH, Maresch, Martin, Khetarpaul, Vipul, Mills, Joseph, Gangwani, Gaurav, Elahwal, Mohamed, Khalil, Rana, Azab, Mohammed A, Mahomed, Anver, Whiston, Richard, Contractor, Ummul, Esposito, Davide, Pratesi, Carlo, Giacomelli, Elena, Troncoso, Martín Veras, Elkouri, Stephane, Johansson, Flavia Gentile, Dodos, Ilias, Benezit, Marie, Vidoedo, José, Rocha-Neves, João, Pereira-Neves, António Henrique, Dias-Neto, Marina Felicidade, Jácome, Ana Filipa Campos, Loureiro, Luis, Silva, Ivone, Garza-Herrera, Rodrigo, Canata, Victor, Bezard, Charlotte, Bowser, Kathryn, Tobar, Jorge Felipe, Vera, Carlos Gomez, Parra, Carolina Salinas, Lopez, Eugenia, Serra, Yvis Gadelha, Varela, Juan, Rubio, Vanessa, Victoria, Gerardo, Johnson, Adam, O’Banion, Leigh Ann, Makar, Ragai, Tantawy, Tamer Ghatwary, Storck, Martin, Jongkind, Vincent, falah, Orwa, McBride, Olivia, Isik, Arda, Papaioannou, Athanasios, Reis, Paulo Eduardo Ocke, Bracale, Umberto Marcello, Atkins, Ellie, Tinelli, Giovanni, Scott, Emma, Wales, Lucy, Sivaharan, Ashwin, Priona, Georgia, Nesbitt, Craig, Grainger, Tabitha, Shelmerdine, Lauren, Chong, Patrick, Bajwa, Adnan, Arwynck, Luke, Hadjievangelou, Nancy, Elbasty, Ahmed, Rubio, Oscar, Ricardo, Michael, Ulloa, Jorge H, Tarazona, Marcos, Pabon, Manuel, Pitoulias, Georgios, and Corless, Kevin
- Subjects
Biomedical and Clinical Sciences ,Cardiovascular Medicine and Haematology ,Clinical Sciences ,Dentistry ,Pediatric Research Initiative ,Prevention ,6.4 Surgery ,Evaluation of treatments and therapeutic interventions ,VERN COVER study collaborative ,AAA ,COVID-19 ,PAD ,Survey ,Vascular surgery ,Cardiovascular medicine and haematology ,Clinical sciences - Abstract
ObjectiveTo document the recovery of vascular services in Europe following the first COVID-19 pandemic peak.MethodsAn online structured vascular service survey with repeated data entry between 23 March and 9 August 2020 was carried out. Unit level data were collected using repeated questionnaires addressing modifications to vascular services during the first peak (March - May 2020, "period 1"), and then again between May and June ("period 2") and June and July 2020 ("period 3"). The duration of each period was similar. From 2 June, as reductions in cases began to be reported, centres were first asked if they were in a region still affected by rising cases, or if they had passed the peak of the first wave. These centres were asked additional questions about adaptations made to their standard pathways to permit elective surgery to resume.ResultsThe impact of the pandemic continued to be felt well after countries' first peak was thought to have passed in 2020. Aneurysm screening had not returned to normal in 21.7% of centres. Carotid surgery was still offered on a case by case basis in 33.8% of centres, and only 52.9% of centres had returned to their normal aneurysm threshold for surgery. Half of centres (49.4%) believed their management of lower limb ischaemia continued to be negatively affected by the pandemic. Reduced operating theatre capacity continued in 45.5% of centres. Twenty per cent of responding centres documented a backlog of at least 20 aortic repairs. At least one negative swab and 14 days of isolation were the most common strategies used for permitting safe elective surgery to recommence.ConclusionCentres reported a broad return of services approaching pre-pandemic "normal" by July 2020. Many introduced protocols to manage peri-operative COVID-19 risk. Backlogs in cases were reported for all major vascular surgeries.
- Published
- 2022
29. Dual Antiplatelet Therapy Does Not Increase Bleeding Risk in Percutaneous Gastrostomy Tube Placement: Network Meta-Analysis
- Author
-
Gangwani, Manesh Kumar, Aziz, Muhammad, Aziz, Abeer, Priyanka, Fnu, Patel, Arti, Ghaffar, Umar, Weissman, Simcha, Asif, Mahmood, Lee-Smith, Wade, Javaid, Toseef, Nawras, Ali, and Hart, Benjamin
- Published
- 2023
- Full Text
- View/download PDF
30. Erythromycin Improves the Quality of Esophagogastroduodenoscopy in Upper Gastrointestinal Bleeding: A Network Meta-Analysis
- Author
-
Aziz, Muhammad, Haghbin, Hossein, Gangwani, Manesh Kumar, Weissman, Simcha, Patel, Arti R., Randhawa, Manraj K., Samikanu, Luke B., Alyousif, Zakaria Abdullah, Lee-Smith, Wade, Kamal, Faisal, Nawras, Ali, and Howden, Colin W.
- Published
- 2023
- Full Text
- View/download PDF
31. Prediction of early‐onset colorectal cancer mortality rates in the United States using machine learning
- Author
-
Hassam Ali, Pratik Patel, Dushyant Singh Dahiya, Manesh Kumar Gangwani, Debargha Basuli, and Babu Pappu Mohan
- Subjects
autoregressive integrated moving average ,cancer prevention ,colorectal cancer ,machine learning ,mortality trends ,Neoplasms. Tumors. Oncology. Including cancer and carcinogens ,RC254-282 - Abstract
Abstract Introduction The current study, focusing on a significant US (United States) colorectal cancer (CRC) burden, employs machine learning for predicting future rates among young population. Methods CDC WONDER data from 1999 to 2022 was analyzed for CRC‐related mortality in patients younger than 56 years. Temporal trends in age‐adjusted mortality rates (AAMRs) were assessed via Joinpoint software. Future mortality rates were forecasted using an optimal Autoregressive Integrated Moving Average (ARIMA) model. Results From 1999 to 2022, we observed 150,908 deaths with CRC listed as the underlying cause, predominantly in males, with an upward trend in AAMR. The ARIMA model projects an increase in CRC mortality by 2035, estimating an average annual percent change (AAPC) of 1.3% overall, 1% for females, and 1.5% for males. Conclusion Our study findings emphasize the need for more robust preventive measures to reduce future CRC mortality among younger population. These results have significant implications for public health policies, particularly for males under 56, and underscore the importance of early screening and lifestyle modifications.
- Published
- 2024
- Full Text
- View/download PDF
32. Harnessing Distribution Ratio Estimators for Learning Agents with Quality and Diversity
- Author
-
Gangwani, Tanmay, Peng, Jian, and Zhou, Yuan
- Subjects
Computer Science - Machine Learning ,Statistics - Machine Learning - Abstract
Quality-Diversity (QD) is a concept from Neuroevolution with some intriguing applications to Reinforcement Learning. It facilitates learning a population of agents where each member is optimized to simultaneously accumulate high task-returns and exhibit behavioral diversity compared to other members. In this paper, we build on a recent kernel-based method for training a QD policy ensemble with Stein variational gradient descent. With kernels based on $f$-divergence between the stationary distributions of policies, we convert the problem to that of efficient estimation of the ratio of these stationary distributions. We then study various distribution ratio estimators used previously for off-policy evaluation and imitation and re-purpose them to compute the gradients for policies in an ensemble such that the resultant population is diverse and of high-quality., Comment: CoRL 2020 camera-ready
- Published
- 2020
33. Learning Guidance Rewards with Trajectory-space Smoothing
- Author
-
Gangwani, Tanmay, Zhou, Yuan, and Peng, Jian
- Subjects
Computer Science - Machine Learning ,Statistics - Machine Learning - Abstract
Long-term temporal credit assignment is an important challenge in deep reinforcement learning (RL). It refers to the ability of the agent to attribute actions to consequences that may occur after a long time interval. Existing policy-gradient and Q-learning algorithms typically rely on dense environmental rewards that provide rich short-term supervision and help with credit assignment. However, they struggle to solve tasks with delays between an action and the corresponding rewarding feedback. To make credit assignment easier, recent works have proposed algorithms to learn dense "guidance" rewards that could be used in place of the sparse or delayed environmental rewards. This paper is in the same vein -- starting with a surrogate RL objective that involves smoothing in the trajectory-space, we arrive at a new algorithm for learning guidance rewards. We show that the guidance rewards have an intuitive interpretation, and can be obtained without training any additional neural networks. Due to the ease of integration, we use the guidance rewards in a few popular algorithms (Q-learning, Actor-Critic, Distributional-RL) and present results in single-agent and multi-agent tasks that elucidate the benefit of our approach when the environmental rewards are sparse or delayed., Comment: NeurIPS 2020 camera-ready
- Published
- 2020
34. Mutual Information Based Knowledge Transfer Under State-Action Dimension Mismatch
- Author
-
Wan, Michael, Gangwani, Tanmay, and Peng, Jian
- Subjects
Statistics - Machine Learning ,Computer Science - Machine Learning - Abstract
Deep reinforcement learning (RL) algorithms have achieved great success on a wide variety of sequential decision-making tasks. However, many of these algorithms suffer from high sample complexity when learning from scratch using environmental rewards, due to issues such as credit-assignment and high-variance gradients, among others. Transfer learning, in which knowledge gained on a source task is applied to more efficiently learn a different but related target task, is a promising approach to improve the sample complexity in RL. Prior work has considered using pre-trained teacher policies to enhance the learning of the student policy, albeit with the constraint that the teacher and the student MDPs share the state-space or the action-space. In this paper, we propose a new framework for transfer learning where the teacher and the student can have arbitrarily different state- and action-spaces. To handle this mismatch, we produce embeddings which can systematically extract knowledge from the teacher policy and value networks, and blend it into the student networks. To train the embeddings, we use a task-aligned loss and show that the representations could be enriched further by adding a mutual information loss. Using a set of challenging simulated robotic locomotion tasks involving many-legged centipedes, we demonstrate successful transfer learning in situations when the teacher and student have different state- and action-spaces., Comment: Conference on Uncertainty in Artificial Intelligence (UAI 2020)
- Published
- 2020
35. State-only Imitation with Transition Dynamics Mismatch
- Author
-
Gangwani, Tanmay and Peng, Jian
- Subjects
Statistics - Machine Learning ,Computer Science - Machine Learning - Abstract
Imitation Learning (IL) is a popular paradigm for training agents to achieve complicated goals by leveraging expert behavior, rather than dealing with the hardships of designing a correct reward function. With the environment modeled as a Markov Decision Process (MDP), most of the existing IL algorithms are contingent on the availability of expert demonstrations in the same MDP as the one in which a new imitator policy is to be learned. This is uncharacteristic of many real-life scenarios where discrepancies between the expert and the imitator MDPs are common, especially in the transition dynamics function. Furthermore, obtaining expert actions may be costly or infeasible, making the recent trend towards state-only IL (where expert demonstrations constitute only states or observations) ever so promising. Building on recent adversarial imitation approaches that are motivated by the idea of divergence minimization, we present a new state-only IL algorithm in this paper. It divides the overall optimization objective into two subproblems by introducing an indirection step and solves the subproblems iteratively. We show that our algorithm is particularly effective when there is a transition dynamics mismatch between the expert and imitator MDPs, while the baseline IL methods suffer from performance degradation. To analyze this, we construct several interesting MDPs by modifying the configuration parameters for the MuJoCo locomotion tasks from OpenAI Gym., Comment: ICLR 2020 camera-ready
- Published
- 2020
36. Fecal microbiota transplant is associated with lower risk of mortality, hepatic encephalopathy, ascites, and infection patients with severe alcohol-associated hepatitis: A systematic review and meta-analysis
- Author
-
Zohaib Ahmed, Andrew Kelly, Joyce Badal, Wade M. Lee-Smith, Manesh Gangwani, Yaseen Alastal, and Mona Hassan
- Subjects
FMT ,Alcoholic Hepatitis ,alcoholism ,Medicine (General) ,R5-920 - Published
- 2023
- Full Text
- View/download PDF
37. Specific and behaviorally consequential astrocyte Gq GPCR signaling attenuation in vivo with iβARK.
- Author
-
Nagai, Jun, Bellafard, Arash, Qu, Zhe, Yu, Xinzhu, Ollivier, Matthias, Gangwani, Mohitkumar R, Diaz-Castro, Blanca, Coppola, Giovanni, Schumacher, Sarah M, Golshani, Peyman, Gradinaru, Viviana, and Khakh, Baljit S
- Subjects
Brain ,Astrocytes ,Neurons ,Animals ,Mice ,Calcium ,Receptors ,G-Protein-Coupled ,Signal Transduction ,beta-Adrenergic Receptor Kinases ,AAV ,GPCR ,Gq ,astrocyte ,behavior ,behavioral adaptation ,calcium ,signaling ,silencing ,spatial memory ,Brain Disorders ,Neurosciences ,Psychology ,Cognitive Sciences ,Neurology & Neurosurgery - Abstract
Astrocytes respond to neurotransmitters and neuromodulators using G-protein-coupled receptors (GPCRs) to mediate physiological responses. Despite their importance, there has been no method to genetically, specifically, and effectively attenuate astrocyte Gq GPCR pathways to explore consequences of this prevalent signaling mechanism in vivo. We report a 122-residue inhibitory peptide from β-adrenergic receptor kinase 1 (iβARK; and inactive D110A control) to attenuate astrocyte Gq GPCR signaling. iβARK significantly attenuated Gq GPCR Ca2+ signaling in brain slices and, in vivo, altered behavioral responses, spared other GPCR responses, and did not alter astrocyte spontaneous Ca2+ signals, morphology, electrophysiological properties, or gene expression in the striatum. Furthermore, brain-wide attenuation of astrocyte Gq GPCR signaling with iβARK using PHP.eB adeno-associated viruses (AAVs), when combined with c-Fos mapping, suggested nuclei-specific contributions to behavioral adaptation and spatial memory. iβARK extends the toolkit needed to explore functions of astrocyte Gq GPCR signaling within neural circuits in vivo.
- Published
- 2021
38. Exploring investor-business-market interplay for business success prediction
- Author
-
Divya Gangwani, Xingquan Zhu, and Borko Furht
- Subjects
Machine learning methods ,Investments-business-market ,Feature engineering ,Success prediction ,Computer engineering. Computer hardware ,TK7885-7895 ,Information technology ,T58.5-58.64 ,Electronic computers. Computer science ,QA75.5-76.95 - Abstract
Abstract The success of the business directly contributes towards the growth of the nation. Hence it is important to evaluate and predict whether the business will be successful or not. In this study, we use the company’s dataset which contains information from startups to Fortune 1000 companies to create a machine learning model for predicting business success. The main challenge of business success prediction is twofold: (1) Identifying variables for defining business success; (2) Feature selection and feature engineering based on Investor-Business-Market interrelation to provide a successful outcome of the predictive modeling. Many studies have been carried out using only the available features to predict business success, however, there is still a challenge to identify the most important features in different business angles and their interrelation with business success. Motivated by the above challenge, we propose a new approach by defining a new business target based on the definition of business success used in this study and develop additional features by carrying out statistical analysis on the training data which highlights the importance of investments, business, and market features in forecasting business success instead of using only the available features for modeling. Ensemble machine learning methods as well as existing supervised learning methods were applied to predict business success. The results demonstrated a significant improvement in the overall accuracy and AUC score using ensemble methods. By adding new features related to the Investor-Business-Market entity demonstrated good performance in predicting business success and proved how important it is to identify significant relationships between these features to cover different business angles when predicting business success. Graphical Abstract
- Published
- 2023
- Full Text
- View/download PDF
39. Correction to: Design Optimization of Water Distribution Networks with Dynamic Search Space Reduction GA
- Author
-
Gangwani, Laxmi, Dongre, Shilpa, Gupta, Rajesh, Sayyed, Mohd Abbas H. Abdy, and Tanyimboh, Tiku
- Published
- 2024
- Full Text
- View/download PDF
40. Evaluating Trust Management Frameworks for Wireless Sensor Networks
- Author
-
Pranav Gangwani, Alexander Perez-Pons, and Himanshu Upadhyay
- Subjects
internet of things (IoT) ,trust management ,wireless sensor networks (WSNs) ,entropy ,beta distribution ,Chemical technology ,TP1-1185 - Abstract
Wireless Sensor Networks (WSNs) are crucial in various fields including Health Care Monitoring, Battlefield Surveillance, and Smart Agriculture. However, WSNs are susceptible to malicious attacks due to the massive quantity of sensors within them. Hence, there is a demand for a trust evaluation framework within WSNs to function as a secure system, to identify and isolate malicious or faulty sensor nodes. This information can be leveraged by neighboring nodes, to prevent collaboration in tasks like data aggregation and forwarding. While numerous trust frameworks have been suggested in the literature to assess trust scores and examine the reliability of sensors through direct and indirect communications, implementing these trust evaluation criteria is challenging due to the intricate nature of the trust evaluation process and the limited availability of datasets. This research conducts a novel comparative analysis of three trust management models: “Lightweight Trust Management based on Bayesian and Entropy (LTMBE)”, “Beta-based Trust and Reputation Evaluation System (BTRES)”, and “Lightweight and Dependable Trust System (LDTS)”. To assess the practicality of these trust management models, we compare and examine their performance in multiple scenarios. Additionally, we assess and compare how well the trust management approaches perform in response to two significant cyber-attacks. Based on the experimental comparative analysis, it can be inferred that the LTMBE model is optimal for WSN applications emphasizing high energy efficiency, while the BTRES model is most suitable for WSN applications prioritizing critical security measures. The conducted empirical comparative analysis can act as a benchmark for upcoming research on trust evaluation frameworks for WSNs.
- Published
- 2024
- Full Text
- View/download PDF
41. Motor adaptation and immediate retention to overground gait-slip perturbation training in people with chronic stroke: an experimental trial with a comparison group
- Author
-
Tanvi Bhatt, Shamali Dusane, Rachana Gangwani, Shuaijie Wang, and Lakshmi Kannan
- Subjects
stroke ,falls ,stability ,adaptation ,reactive balance ,Sports ,GV557-1198.995 - Abstract
BackgroundPerturbation-based training has shown to be effective in reducing fall-risk in people with chronic stroke (PwCS). However, most evidence comes from treadmill-based stance studies, with a lack of research focusing on training overground perturbed walking and exploring the relative contributions of the paretic and non-paretic limbs. This study thus examined whether PwCS could acquire motor adaptation and demonstrate immediate retention of fall-resisting skills following bilateral overground gait-slip perturbation training.Methods65 PwCS were randomly assigned to either (i) a training group, that received blocks of eight non-paretic (NP-S1 to NP-S8) and paretic (P-S1 to P-S8) overground slips during walking followed by a mixed block (seven non-paretic and paretic slips each interspersed with unperturbed walking trials) (NP-S9/P-S9 to NP-S15/P-S15) or (ii) a control group, that received a single non-paretic and paretic slip in random order. The assessor and training personnel were not blinded. Immediate retention was tested for the training group after a 30-minute rest break. Primary outcomes included laboratory-induced slip outcomes (falls and balance loss) and center of mass (CoM) state stability. Secondary outcomes to understand kinematic contributors to stability included recovery strategies, limb kinematics, slipping kinematics, and recovery stride length.ResultsPwCS within the training group showed reduced falls (p 0.01). On comparing the first and last training trial (S1 vs. S15), post-slip stability improved on both non-paretic and paretic slips, however, pre-slip stability improved only on the non-paretic slip (p
- Published
- 2023
- Full Text
- View/download PDF
42. Learning Belief Representations for Imitation Learning in POMDPs
- Author
-
Gangwani, Tanmay, Lehman, Joel, Liu, Qiang, and Peng, Jian
- Subjects
Computer Science - Machine Learning ,Statistics - Machine Learning - Abstract
We consider the problem of imitation learning from expert demonstrations in partially observable Markov decision processes (POMDPs). Belief representations, which characterize the distribution over the latent states in a POMDP, have been modeled using recurrent neural networks and probabilistic latent variable models, and shown to be effective for reinforcement learning in POMDPs. In this work, we investigate the belief representation learning problem for generative adversarial imitation learning in POMDPs. Instead of training the belief module and the policy separately as suggested in prior work, we learn the belief module jointly with the policy, using a task-aware imitation loss to ensure that the representation is more aligned with the policy's objective. To improve robustness of representation, we introduce several informative belief regularization techniques, including multi-step prediction of dynamics and action-sequences. Evaluated on various partially observable continuous-control locomotion tasks, our belief-module imitation learning approach (BMIL) substantially outperforms several baselines, including the original GAIL algorithm and the task-agnostic belief learning algorithm. Extensive ablation analysis indicates the effectiveness of task-aware belief learning and belief regularization., Comment: Conference on Uncertainty in Artificial Intelligence (UAI 2019)
- Published
- 2019
43. Partial Redundancy Elimination using Lazy Code Motion
- Author
-
Dasgupta, Sandeep and Gangwani, Tanmay
- Subjects
Computer Science - Programming Languages - Abstract
Partial Redundancy Elimination (PRE) is a compiler optimization that eliminates expressions that are redundant on some but not necessarily all paths through a program. In this project, we implemented a PRE optimization pass in LLVM and measured results on a variety of applications. We chose PRE because it is a powerful technique that subsumes Common Subexpression Elimination (CSE) and Loop Invariant Code Motion (LICM), and hence has the potential to greatly improve performance.
- Published
- 2019
44. A Blockchain-Facilitated Secure Sensing Data Processing and Logging System
- Author
-
Wenbing Zhao, Izdehar M. Aldyaflah, Pranav Gangwani, Santosh Joshi, Himanshu Upadhyay, and Leonel Lagos
- Subjects
Blockchain ,cyber-physical systems ,data immutability ,data processing and logging ,distributed ledger ,Merkle tree ,Electrical engineering. Electronics. Nuclear engineering ,TK1-9971 - Abstract
In this paper, we present the design, implementation, and evaluation of a secure sensing data processing and logging system. The system is inspired and enabled by the blockchain technology. In this system, a public blockchain is used as immutable datastore to log the most critical data needed to secure the system. Furthermore, several innovative blockchain-inspired mechanisms have been incorporated into the system to provide additional security for the system’s operations. The first priority in securing sensing data processing and logging is admission control, i.e., only legitimate sensing data are accepted for processing and logging. This is achieved via a sensor identification and authentication mechanism. The second priority is to ensure that the logged data remain intact overtime. This is achieved by storing a small amount of data condensed from the raw sensing data on a public blockchain. A Merkel-tree based mechanism is devised to link the raw sensing data stored off-chain to the condensed data placed on public blockchain. This mechanism passes the data immutability property of a public blockchain to the raw sensing data stored off-chain. Third, the raw sensing data stored off-chain are secured with a self-protection mechanism where the raw sensing data are grouped into chained blocks with a moderate amount of proof-of-work. This scheme prevents an adversary from making arbitrary changes to the logged data within a short period of time. Fourth, mechanisms are developed to facilitate the search of the condensed data placed on the public blockchain and the verification of the raw sensing data using the condensed data placed on the public blockchain. The system is implemented in Python except the graphical user interface, which is developed using C#. The functionality and feasibility of the system have been evaluated locally and with two public blockchain systems, one is the IOTA Shimmer test network, and the other is Ethereum.
- Published
- 2023
- Full Text
- View/download PDF
45. Does a Compatibilizer Enhance the Properties of Carbon Fiber-Reinforced Composites?
- Author
-
Prashant Gangwani, Mitjan Kalin, and Nazanin Emami
- Subjects
compatibilizer ,carbon fiber ,polymer composites ,UHMWPE ,PTFE ,PPS ,Organic chemistry ,QD241-441 - Abstract
We have evaluated the effectiveness of compatibilizers in blends and composites produced using a solvent manufacturing process. The compatibilizers were two different types of polyethylene (linear low-density and high-density) grafted with maleic anhydride (MAH) and a highly functionalized, epoxy-based compatibilizer with the tradename Joncryl. The selected material combinations were an ultra-high-molecular-weight polyethylene (UHMWPE) with MAH-based materials as compatibilizers and a polyphenylene sulfide plus polytetrafluoroethylene (PPS-PTFE) polymer blend with an epoxy-based compatibilizer. The findings revealed that while the compatibilizers consistently enhanced the properties, such as the impact strength and hardness of PPS-based compositions, their utility is constrained to less complex compositions, such as fibrous-reinforced PPS or PPS-PTFE polymer blends. For fibrous-reinforced PPS-PTFE composites, the improvement in performance does not justify the presence of compatibilizers. In contrast, for UHMWPE compositions, compatibilizers demonstrated negligible or even detrimental effects, particularly in reinforced UHMWPE. Overall, the epoxy-based compatibilizer Joncryl stands out as the only effective option for enhancing mechanical performance. Thermal and chemical characterization indicated that the compatibilizers function as chain extenders and enhance the fiber–matrix interface in PPS-based compositions, while they remain inactive in UHMWPE-based compositions. Ultimately, the incompatibility of the compatibilizers with certain aspects of the manufacturing method and the inconsistent integration with the polymer are the main reasons for their ineffectiveness in UHMWPE compositions.
- Published
- 2023
- Full Text
- View/download PDF
46. Distributed and Secure ML with Self-tallying Multi-party Aggregation
- Author
-
Long, Yunhui, Gangwani, Tanmay, Mughees, Haris, and Gunter, Carl
- Subjects
Computer Science - Cryptography and Security - Abstract
Privacy preserving multi-party computation has many applications in areas such as medicine and online advertisements. In this work, we propose a framework for distributed, secure machine learning among untrusted individuals. The framework consists of two parts: a two-step training protocol based on homomorphic addition and a zero knowledge proof for data validity. By combining these two techniques, our framework provides privacy of per-user data, prevents against a malicious user contributing corrupted data to the shared pool, enables each user to self-compute the results of the algorithm without relying on external trusted third parties, and requires no private channels between groups of users. We show how different ML algorithms such as Latent Dirichlet Allocation, Naive Bayes, Decision Trees etc. fit our framework for distributed, secure computing., Comment: NeurIPS 2018 Workshop on PPML
- Published
- 2018
47. Learning Self-Imitating Diverse Policies
- Author
-
Gangwani, Tanmay, Liu, Qiang, and Peng, Jian
- Subjects
Statistics - Machine Learning ,Computer Science - Machine Learning - Abstract
The success of popular algorithms for deep reinforcement learning, such as policy-gradients and Q-learning, relies heavily on the availability of an informative reward signal at each timestep of the sequential decision-making process. When rewards are only sparsely available during an episode, or a rewarding feedback is provided only after episode termination, these algorithms perform sub-optimally due to the difficultly in credit assignment. Alternatively, trajectory-based policy optimization methods, such as cross-entropy method and evolution strategies, do not require per-timestep rewards, but have been found to suffer from high sample complexity by completing forgoing the temporal nature of the problem. Improving the efficiency of RL algorithms in real-world problems with sparse or episodic rewards is therefore a pressing need. In this work, we introduce a self-imitation learning algorithm that exploits and explores well in the sparse and episodic reward settings. We view each policy as a state-action visitation distribution and formulate policy optimization as a divergence minimization problem. We show that with Jensen-Shannon divergence, this divergence minimization problem can be reduced into a policy-gradient algorithm with shaped rewards learned from experience replays. Experimental results indicate that our algorithm works comparable to existing algorithms in environments with dense rewards, and significantly better in environments with sparse and episodic rewards. We then discuss limitations of self-imitation learning, and propose to solve them by using Stein variational policy gradient descent with the Jensen-Shannon kernel to learn multiple diverse policies. We demonstrate its effectiveness on a challenging variant of continuous-control MuJoCo locomotion tasks., Comment: ICLR 2019
- Published
- 2018
48. Hyperactivity with Disrupted Attention by Activation of an Astrocyte Synaptogenic Cue.
- Author
-
Nagai, Jun, Rajbhandari, Abha, Gangwani, Mohitkumar, Hachisuka, Ayaka, Coppola, Giovanni, Khakh, Baljit, Masmanidis, Sotirios, and Fanselow, Michael
- Subjects
astrocyte ,attention deficit ,behavior ,calcium ,gabapentin ,hyperactivity ,microcircuit ,striatum ,thrombospondin ,Animals ,Astrocytes ,Attention Deficit Disorder with Hyperactivity ,Behavior ,Animal ,Cell Communication ,Female ,Male ,Mice ,Mice ,Transgenic ,Neurons ,Receptors ,GABA-B ,Signal Transduction ,Synapses ,Thrombospondin 1 ,gamma-Aminobutyric Acid - Abstract
Hyperactivity and disturbances of attention are common behavioral disorders whose underlying cellular and neural circuit causes are not understood. We report the discovery that striatal astrocytes drive such phenotypes through a hitherto unknown synaptic mechanism. We found that striatal medium spiny neurons (MSNs) triggered astrocyte signaling via γ-aminobutyric acid B (GABAB) receptors. Selective chemogenetic activation of this pathway in striatal astrocytes in vivo resulted in acute behavioral hyperactivity and disrupted attention. Such responses also resulted in upregulation of the synaptogenic cue thrombospondin-1 (TSP1) in astrocytes, increased excitatory synapses, enhanced corticostriatal synaptic transmission, and increased MSN action potential firing in vivo. All of these changes were reversed by blocking TSP1 effects. Our data identify a form of bidirectional neuron-astrocyte communication and demonstrate that acute reactivation of a single latent astrocyte synaptogenic cue alters striatal circuits controlling behavior, revealing astrocytes and the TSP1 pathway as therapeutic targets in hyperactivity, attention deficit, and related psychiatric disorders.
- Published
- 2019
49. Transient, Consequential Increases in Extracellular Potassium Ions Accompany Channelrhodopsin2 Excitation
- Author
-
Octeau, J Christopher, Gangwani, Mohitkumar R, Allam, Sushmita L, Tran, Duy, Huang, Shuhan, Hoang-Trong, Tuan M, Golshani, Peyman, Rumbell, Timothy H, Kozloski, James R, and Khakh, Baljit S
- Subjects
Biological Sciences ,Neurosciences ,Genetics ,Neurological ,Animals ,Channelrhodopsins ,Mice ,Optogenetics ,Potassium ,astrocyte ,channelrhodopsin ,circuit ,neuron ,optogenetics ,potassium ,striatum ,Biochemistry and Cell Biology ,Medical Physiology ,Biological sciences - Abstract
Channelrhodopsin2 (ChR2) optogenetic excitation is widely used to study neurons, astrocytes, and circuits. Using complementary approaches in situ and in vivo, we found that ChR2 stimulation leads to significant transient elevation of extracellular potassium ions by ∼5 mM. Such elevations were detected in ChR2-expressing mice, following local in vivo expression of ChR2(H134R) with adeno-associated viruses (AAVs), in different brain areas and when ChR2 was expressed in neurons or astrocytes. In particular, ChR2-mediated excitation of striatal astrocytes was sufficient to increase medium spiny neuron (MSN) excitability and immediate early gene expression. The effects on MSN excitability were recapitulated in silico with a computational MSN model and detected in vivo as increased action potential firing in awake, behaving mice. We show that transient, physiologically consequential increases in extracellular potassium ions accompany ChR2 optogenetic excitation. This coincidental effect may be important to consider during astrocyte studies employing ChR2 to interrogate neural circuits and animal behavior.
- Published
- 2019
50. Efficacy and Safety of EUS-directed Transgastric ERCP (EDGE) vs Laparoscopic-Assisted ERCP
- Author
-
Manesh Gangwani
- Subjects
Hospital Medicine ,Research ,Medicine (General) ,R5-920 - Abstract
Please view the PDF to see the formatted meeting abstract.
- Published
- 2023
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.