scholarly journals A Deep Gravity model for mobility flows generation

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Filippo Simini ◽  
Gianni Barlacchi ◽  
Massimilano Luca ◽  
Luca Pappalardo
Keyword(s):  
Neural Networks ◽  
New York ◽  
Gravity Model ◽  
Deep Neural Networks ◽  
New York State ◽  
Region Of Interest ◽  
Well Being ◽  
Data Availability ◽  
Linear Relationships ◽  
Geographic Data

AbstractThe movements of individuals within and among cities influence critical aspects of our society, such as well-being, the spreading of epidemics, and the quality of the environment. When information about mobility flows is not available for a particular region of interest, we must rely on mathematical models to generate them. In this work, we propose Deep Gravity, an effective model to generate flow probabilities that exploits many features (e.g., land use, road network, transport, food, health facilities) extracted from voluntary geographic data, and uses deep neural networks to discover non-linear relationships between those features and mobility flows. Our experiments, conducted on mobility flows in England, Italy, and New York State, show that Deep Gravity achieves a significant increase in performance, especially in densely populated regions of interest, with respect to the classic gravity model and models that do not use deep neural networks or geographic data. Deep Gravity has good generalization capability, generating realistic flows also for geographic areas for which there is no data availability for training. Finally, we show how flows generated by Deep Gravity may be explained in terms of the geographic features and highlight crucial differences among the three considered countries interpreting the model’s prediction with explainable AI techniques.

“There Was No ‘That’s Not My Job’”: New York Area Agencies on Aging Approaches to Supporting Older Adults During the COVID-19 Pandemic

2021 ◽  
pp. 073346482199102
Author(s):  
Claire Pendergrast
Keyword(s):  
Older Adults ◽  
New York ◽  
New York State ◽  
Well Being ◽  
Contextual Influences ◽  
Structured Interviews ◽  
Response And Recovery ◽  
Pandemic Response ◽  
Increasing Demand ◽  
Priority Needs

The COVID-19 pandemic has disrupted many older adults’ traditional sources of formal and informal supports, increasing demand for Area Agency on Aging services (AAAs). This study examines strategies used by AAAs to support older adults’ health and well-being during COVID-19 and identifies contextual influences on AAA pandemic response activities. Semi-structured interviews were conducted with representatives of 20 AAAs in New York State. A combined inductive and deductive approach was used to code and thematically analyze the data. AAAs rapidly expanded capacity and dramatically modified program offerings, communications activities, and service delivery protocols to address emergent needs and minimize COVID-19 exposure risk for clients. AAAs’ trusted relationships with older adults and community partners improved their capacity to identify priority needs and coordinate appropriate supports. Policymakers should ensure that AAAs receive sustained financial and technical support to ensure critical community-based services are available for older adults throughout pandemic response and recovery.

scholarly journals Dynamics and future of SARS-CoV-2 in the human host

2020 ◽  
Author(s):  
Michael Gillman ◽  
Nuno Crokidakis
Keyword(s):  
New York ◽  
Global Economy ◽  
New York State ◽  
Well Being ◽  
Logistic Growth ◽  
Virus Growth ◽  
Human Host ◽  
Population Dynamics Model ◽  
National Governments ◽  
Mental Well Being

Forecasting trends in COVID-19 infections is vital for the global economy, national governments and physical and mental well-being. Using the per capita number of new cases as a proxy for the abundance of the SARS-CoV-2 virus, and the number of deaths as a measure of virulence, the dynamics of the pandemic and the outcomes emerging from it are examined for three locations (England, Italy and New York State). The data are analysed with a new version of a population dynamics model that combines exponential/logistic growth with time-varying carrying capacity, allowing predictions of persistence or extinction of the virus. In agreement with coevolutionary theory, the model suggests a transition from exponential virus growth to low abundance, coupled with reduced virulence, during colonisation of the alternate human host. The structure of the model allows a straightforward assessment of key parameters, which can be contrasted with standard epidemiological models and interpreted with respect to ecological and evolutionary processes and isolation policies.

scholarly journals Integrating Deep Neural Networks and Symbolic Inference for Organic Reactivity Prediction

2020 ◽  
Author(s):  
Wesley Wei Qian ◽  
Nathan T. Russell ◽  
Claire L. W. Simons ◽  
Yunan Luo ◽  
Martin D. Burke ◽  
...  
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Probability Distribution ◽  
Deep Neural Networks ◽  
Machine Intelligence ◽  
Data Availability ◽  
Learning Models ◽  
Expert Performance ◽  
Covalent Bonds ◽  
Chemical Reasoning

<div>Accurate <i>in silico</i> models for the prediction of novel chemical reaction outcomes can be used to guide the rapid discovery of new reactivity and enable novel synthesis strategies for newly discovered lead compounds. Recent advances in machine learning, driven by deep learning models and data availability, have shown utility throughout synthetic organic chemistry as a data-driven method for reaction prediction. Here we present a machine-intelligence approach to predict the products of an organic reaction by integrating deep neural networks with a probabilistic and symbolic inference that flexibly enforces chemical constraints and accounts for prior chemical knowledge. We first train a graph convolutional neural network to estimate the likelihood of changes in covalent bonds, hydrogen counts, and formal charges. These estimated likelihoods govern a probability distribution over potential products. Integer Linear Programming is then used to infer the most probable products from the probability distribution subject to heuristic rules such as the octet rule and chemical constraints that reflect a user's prior knowledge. Our approach outperforms previous graph-based neural networks by predicting products with more than 90% accuracy, demonstrates intuitive chemical reasoning through a learned attention mechanism, and provides generalizability across various reaction types. Furthermore, we demonstrate the potential for even higher model accuracy when complemented by expert chemists contributing to the system, boosting both machine and expert performance. The results show the advantages of empowering deep learning models with chemical intuition and knowledge to expedite the drug discovery process.</div>

scholarly journals Coping Through COVID-19: A Mixed Method Approach to Understand How Palliative Care Teams Managed the COVID-19 Pandemic

2021 ◽  
pp. 104990912110456
Author(s):  
Kayla D. Finuf ◽  
Santiago Lopez ◽  
Maria T. Carney
Keyword(s):  
New York ◽  
Palliative Care ◽  
Physical Health ◽  
Coping Strategies ◽  
New York State ◽  
Supervisor Support ◽  
Well Being ◽  
Subjective Well Being ◽  
Positive Effects ◽  
Palliative Care Teams

Objective: While previous work documented a substantial increase in patient mortality consultations and workload for palliative teams, little is known about how these team members managed their mental and physical health during the COVID-19 pandemic. We investigated how job resources (coworker and supervisor support) and personal resources (coping strategies) reduced perceptions of burnout and increased perceptions of well-being. Method: An anonymous electronic survey was sent to all members ( N = 64) of the palliative medical team among 14 hospitals of a New York State health system. Data were collected between September 2020 to October 2020. Measures included validated scales for burnout (Oldenburg Burnout Inventory), coping strategies (Cybernetic Coping Scale), subjective well-being (BBC Subjective Well-being scale), and coworker/supervisor support (7 items from Yang et al). Results: Results indicated devaluation coping tactics were used to reduce perceptions of burnout and to increase perceptions of physical health. Higher burnout was identified when using avoidance coping techniques. Furthermore, coworkers and supervisor(s) support significantly reduced disengagement when compared to coworker support alone. Conclusion: COVID-19 exacerbated burnout experienced by palliative care teams, yet the use of coping behaviors (devaluation/avoidance) and external resources (coworker and supervisor support) utilized by these teams were found to have positive effects. Further research should investigate these antagonizing factors to help preventing and addressing burn out during times of crises and in the everyday of palliative care teams.

scholarly journals Raabin-WBC: a large free access dataset of white blood cells from normal peripheral blood

2021 ◽  
Author(s):  
Seyedeh-Zahra Mousavi Kouzehkanan ◽  
Sepehr Saghari ◽  
Eslam Tavakoli ◽  
Peyman Rostami ◽  
Mohammadjavad AbbasZadeh ◽  
...  
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Peripheral Blood ◽  
Blood Cells ◽  
Deep Neural Networks ◽  
New Technologies ◽  
White Blood Cells ◽  
Well Being ◽  
Free Access ◽  
Normal Peripheral Blood

Accurate and early detection of peripheral white blood cell anomalies plays a crucial role in the evaluation of an individual's well-being. The emergence of new technologies such as artificial intelligence can be very effective in achieving this. In this regard, most of the state-of-the-art methods use deep neural networks. Data can significantly influence the performance and generalization power of machine learning approaches, especially deep neural networks. To that end, we collected a large free available dataset of white blood cells from normal peripheral blood samples called Raabin-WBC. Our dataset contains about 40000 white blood cells and artifacts (color spots). To reassure correct data, a significant number of cells were labeled by two experts, and the ground truth of nucleus and cytoplasm were extracted by experts for some cells (about 1145), as well. To provide the necessary diversity, various smears have been imaged. Hence, two different cameras and two different microscopes were used. The Raabin-WBC dataset can be used for different machine learning tasks such as classification, detection, segmentation, and localization. We also did some primary deep learning experiments on Raabin-WBC, and we showed how the generalization power of machine learning methods, especially deep neural networks, was affected by the mentioned diversity.

scholarly journals Dynamic Ride-Hailing with Electric Vehicles

2021 ◽  
Author(s):  
Nicholas D. Kullman ◽  
Martin Cousineau ◽  
Justin C. Goodson ◽  
Jorge E. Mendoza
Keyword(s):  
Neural Networks ◽  
Decision Making ◽  
New York ◽  
New York City ◽  
Reinforcement Learning ◽  
Electric Vehicles ◽  
Optimal Policy ◽  
Deep Neural Networks ◽  
Real Data ◽  
Dual Bounds

We consider the problem of an operator controlling a fleet of electric vehicles for use in a ride-hailing service. The operator, seeking to maximize profit, must assign vehicles to requests as they arise as well as recharge and reposition vehicles in anticipation of future requests. To solve this problem, we employ deep reinforcement learning, developing policies whose decision making uses [Formula: see text]-value approximations learned by deep neural networks. We compare these policies against a reoptimization-based policy and against dual bounds on the value of an optimal policy, including the value of an optimal policy with perfect information, which we establish using a Benders-based decomposition. We assess performance on instances derived from real data for the island of Manhattan in New York City. We find that, across instances of varying size, our best policy trained with deep reinforcement learning outperforms the reoptimization approach. We also provide evidence that this policy may be effectively scaled and deployed on larger instances without retraining.

scholarly journals Raabin-WBC: a large free access dataset of white blood cells from normal peripheral blood

2021 ◽  
Author(s):  
Zahra Mousavi Kouzehkanan ◽  
Sepehr Saghari ◽  
Eslam Tavakoli ◽  
Peyman Rostami ◽  
Mohammadjavad Abaszadeh ◽  
...  
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Peripheral Blood ◽  
Blood Cells ◽  
Deep Neural Networks ◽  
New Technologies ◽  
White Blood Cells ◽  
Well Being ◽  
Free Access ◽  
Normal Peripheral Blood

Abstract Accurate and early detection of peripheral white blood cell anomalies plays a crucial role in the evaluation of an individual's well-being. The emergence of new technologies such as artificial intelligence can be very effective in achieving this. In this regard, most of the state-of-the-art methods use deep neural networks. Data can significantly influence the performance and generalization power of machine learning approaches, especially deep neural networks. To that end, we collected a large free available dataset of white blood cells from normal peripheral blood samples called Raabin-WBC. Our dataset contains about 40000 white blood cells and artifacts (color spots). To reassure correct data, a significant number of cells were labeled by two experts, and the ground truth of nucleus and cytoplasm were extracted by experts for some cells (about 1145), as well. To provide the necessary diversity, various smears have been imaged. Hence, two different cameras and two different microscopes were used. The Raabin-WBC dataset can be used for different machine learning tasks such as classification, detection, segmentation, and localization. We also did some primary deep learning experiments on Raabin-WBC, and we showed how the generalization power of machine learning methods, especially deep neural networks, was affected by the mentioned diversity.

scholarly journals PHENOLOGY OF THE SAN JOSE SCALE (HOMOPTERA: DIASPIDIDAE) IN NEW YORK STATE APPLE ORCHARDS,

1983 ◽  
Vol 115 (7) ◽  
pp. 717-722 ◽  
Author(s):  
Diane L. Mague ◽  
W. Harvey Reissig
Keyword(s):  
New York ◽  
New York State ◽  
Pheromone Trap ◽  
Apple Orchards ◽  
San Jose ◽  
Degree Days ◽  
Pheromone Trapping ◽  
Linear Relationships ◽  
San Jose Scale ◽  
Trap Catches

AbstractPheromone trapping studies from 1979 to 1981 showed that there were two periods of San Jose scale, Quadraspidiotus perniciosus (Comstock), male flight activity annually in western New York apple orchards. Spring flight, which resulted from overwintering black caps, began at ca. 94–140 degree-days (base 10 °C from 1 March) and occurred during bloom in the apple varieties studied. First generation crawlers emerged at ca. 360 degree-days. Second generation crawlers emerged at ca. 890 degree-days and were active throughout September. Regression analyses showed a logistic relationship between crawler density and fruit infestation, and inverse linear relationships between pheromone trap catches and San Jose scale infestation levels within trees.

Integrating Deep Neural Networks and Symbolic Inference for Organic Reactivity Prediction

2020 ◽  
Author(s):  
Wesley Wei Qian ◽  
Nathan T. Russell ◽  
Claire L. W. Simons ◽  
Yunan Luo ◽  
Martin D. Burke ◽  
...  
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Probability Distribution ◽  
Deep Neural Networks ◽  
Machine Intelligence ◽  
Data Availability ◽  
Learning Models ◽  
Expert Performance ◽  
Covalent Bonds ◽  
Chemical Reasoning

<div>Accurate <i>in silico</i> models for the prediction of novel chemical reaction outcomes can be used to guide the rapid discovery of new reactivity and enable novel synthesis strategies for newly discovered lead compounds. Recent advances in machine learning, driven by deep learning models and data availability, have shown utility throughout synthetic organic chemistry as a data-driven method for reaction prediction. Here we present a machine-intelligence approach to predict the products of an organic reaction by integrating deep neural networks with a probabilistic and symbolic inference that flexibly enforces chemical constraints and accounts for prior chemical knowledge. We first train a graph convolutional neural network to estimate the likelihood of changes in covalent bonds, hydrogen counts, and formal charges. These estimated likelihoods govern a probability distribution over potential products. Integer Linear Programming is then used to infer the most probable products from the probability distribution subject to heuristic rules such as the octet rule and chemical constraints that reflect a user's prior knowledge. Our approach outperforms previous graph-based neural networks by predicting products with more than 90% accuracy, demonstrates intuitive chemical reasoning through a learned attention mechanism, and provides generalizability across various reaction types. Furthermore, we demonstrate the potential for even higher model accuracy when complemented by expert chemists contributing to the system, boosting both machine and expert performance. The results show the advantages of empowering deep learning models with chemical intuition and knowledge to expedite the drug discovery process.</div>

Sign in / Sign up

Export Citation Format

Share Document