Author name cluster

Sahil Sharma

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

2 papers

1 author row

JBHI Journal 2023 Journal Article

SwasthGarbh: A Smartphone App for Improving the Quality of Antenatal Care and Ameliorating Maternal-Fetal Health

Sahil Sharma
Snigdha Soni
Shalini Kaushik
Mani Kalaivani
Vatsla Dadhwal
K Aparna Sharma
Deepak Sharma

The utility of telemedicine in healthcare has been brought to the forefront by the COVID-19 pandemic. ‘SwasthGarbh’ (Healthy Pregnancy) is a multi-functional, interactive smartphone application for providing antenatal care and real-time medical support to all pregnant women (especially those in rural areas and/or do not have easy access to doctors). A randomized controlled trial ( n = 150) demonstrates its utility in improving the quality of antenatal care, reducing obstetric/medical complications and achieving a positive pregnancy experience. The test group (patients registered on the App) showed a significantly higher number of mean (± SD) antenatal visits (7. 0 ± 1. 5 vs. 5. 7 ± 1. 8; P vs. 69. 8%, P vs. 41. 7%, P vs. 55. 5%, P = 0. 04) and obstetric (52. 1% vs. 59. 7%, P = 0. 36) complications during pregnancy as well as significant improvement in mean (± SD) maternal systolic BP (118. 9 ± 11. 8 vs. 123. 4 ± 14. 2 mmHg; P = 0. 046), diastolic BP (76. 0 ± 8. 4 vs. 80. 0 ± 10. 9 mmHg; P = 0. 02) and hemoglobin (11. 5 ± 1. 4 vs. 10. 9 ± 1. 4 g/dL; P = 0. 03) parameters at delivery was observed in the test group compared to the controls, respectively. All the above mentioned positive clinical outcomes were the result of the provision of high quality antenatal care, timely detection of complications, prompt medical assistance and improved medication adherence. This is first pregnancy App that provides instantaneous access to doctor's advice and is clinically endorsed as well as credible.

Details DOI

AAAI Conference 2017 Conference Paper

Dynamic Action Repetition for Deep Reinforcement Learning

Aravind Lakshminarayanan
Sahil Sharma
Balaraman Ravindran

One of the long standing goals of Artiﬁcial Intelligence (AI) is to build cognitive agents which can perform complex tasks from raw sensory inputs without explicit supervision (Lake et al. 2016). Recent progress in combining Reinforcement Learning objective functions and Deep Learning architectures has achieved promising results for such tasks. An important aspect of such sequential decision making problems, which has largely been neglected, is for the agent to decide on the duration of time for which to commit to actions. Such action repetition is important for computational efﬁciency, which is necessary for the agent to respond in real-time to events (in applications such as self-driving cars). Action Repetition arises naturally in real life as well as simulated environments. The time scale of executing an action enables an agent (both humans and AI) to decide the granularity of control during task execution. Current state of the art Deep Reinforcement Learning models, whether they are off-policy (Mnih et al. 2015; Wang et al. 2015) or on-policy (Mnih et al. 2016), consist of a framework with a static action repetition paradigm, wherein the action decided by the agent is repeated for a ﬁxed number of time steps regardless of the contextual state while executing the task. In this paper, we propose a new framework - Dynamic Action Repetition which changes Action Repetition Rate (the time scale of repeating an action) from a hyper-parameter of an algorithm to a dynamically learnable quantity. At every decision-making step, our models allow the agent to commit to an action and the time scale of executing the action. We show empirically that such a dynamic time scale mechanism improves the performance on relatively harder games in the Atari 2600 domain, independent of the underlying Deep Reinforcement Learning algorithm used.

PDF Details