make edits to citations in the paper

paper/paper.bib

@@ -3,7 +3,6 @@ @misc{towers_gymnasium_2023
     url = {https://zenodo.org/record/8127025},
     abstract = {An API standard for single-agent reinforcement learning environments, with popular reference environments and related utilities (formerly Gym)},
     urldate = {2023-07-08},
-    publisher = {Zenodo},
     author = {Towers, Mark and Terry, Jordan K. and Kwiatkowski, Ariel and Balis, John U. and Cola, Gianluca de and Deleu, Tristan and Goulão, Manuel and Kallinteris, Andreas and KG, Arjun and Krimmel, Markus and Perez-Vicente, Rodrigo and Pierré, Andrea and Schulhoff, Sander and Tai, Jun Jet and Shen, Andrew Tan Jin and Younis, Omar G.},
     month = mar,
     year = {2023},
@@ -39,7 +38,7 @@ @article{https://doi.org/10.1002/eap.1898
     number = {6},
     pages = {e01898},
     keywords = {Bayesian, climate, extremes, fire, spatiotemporal, wildfire},
-    doi = {https://doi.org/10.1002/eap.1898},
+    doi = {10.1002/eap.1898},
     url = {https://esajournals.onlinelibrary.wiley.com/doi/abs/10.1002/eap.1898},
     eprint = {https://esajournals.onlinelibrary.wiley.com/doi/pdf/10.1002/eap.1898},
     abstract = {Abstract Wildfires are becoming more frequent in parts of the globe, but predicting where and when wildfires occur remains difficult. To predict wildfire extremes across the contiguous United States, we integrate a 30-yr wildfire record with meteorological and housing data in spatiotemporal Bayesian statistical models with spatially varying nonlinear effects. We compared different distributions for the number and sizes of large fires to generate a posterior predictive distribution based on finite sample maxima for extreme events (the largest fires over bounded spatiotemporal domains). A zero-inflated negative binomial model for fire counts and a lognormal model for burned areas provided the best performance. This model attains 99\% interval coverage for the number of fires and 93\% coverage for fire sizes over a six year withheld data set. Dryness and air temperature strongly predict extreme wildfire probabilities. Housing density has a hump-shaped relationship with fire occurrence, with more fires occurring at intermediate housing densities. Statistically, these drivers affect the chance of an extreme wildfire in two ways: by altering fire size distributions, and by altering fire frequency, which influences sampling from the tails of fire size distributions. We conclude that recent extremes should not be surprising, and that the contiguous United States may be on the verge of even larger wildfire extremes.},
@@ -54,17 +53,17 @@ @article{diao2020uncertainty
 }
 
 @article{ganapathi2018using,
-  title={Using spatial reinforcement learning to build forest wildfire dynamics models from satellite images},
+  title={{Using spatial reinforcement learning to build forest wildfire dynamics models from satellite images}},
   author={Ganapathi Subramanian, Sriram and Crowley, Mark},
-  journal={Frontiers in ICT},
+  journal={{Frontiers in Information and Communication Technology}},
   volume={5},
   pages={6},
   year={2018},
   publisher={Frontiers Media SA}
 }
 
 @article{julian2019distributed,
-  title={Distributed wildfire surveillance with autonomous aircraft using deep reinforcement learning},
+  title={{Distributed wildfire surveillance with autonomous aircraft using deep reinforcement learning}},
   author={Julian, Kyle D and Kochenderfer, Mykel J},
   journal={Journal of Guidance, Control, and Dynamics},
   volume={42},
@@ -75,7 +74,7 @@ @article{julian2019distributed
 }
 
 @article{altamimi2022large,
-  title={Large-Scale Wildfire Mitigation Through Deep Reinforcement Learning},
+  title={{Large-Scale Wildfire Mitigation Through Deep Reinforcement Learning}},
   author={Altamimi, Abdulelah and Lagoa, Constantino and Borges, Jos{\'e} G and McDill, Marc E and Andriotis, CP and Papakonstantinou, KG},
   journal={Frontiers in Forests and Global Change},
   volume={5},
@@ -87,7 +86,7 @@ @article{altamimi2022large
 @ARTICLE{9340340,
   author={Viseras, Alberto and Meissner, Michael and Marchal, Juan},
   journal={IEEE Access},
-  title={Wildfire Front Monitoring with Multiple UAVs using Deep Q-Learning},
+  title={{Wildfire Front Monitoring with Multiple UAVs using Deep Q-Learning}},
   year={2021},
   volume={},
   number={},
@@ -103,7 +102,7 @@ @article{https://doi.org/10.1111/risa.12944
     number = {7},
     pages = {1390-1404},
     keywords = {Decision making, evacuation, risk attitudes, wildfires},
-    doi = {https://doi.org/10.1111/risa.12944},
+    doi = {10.1111/risa.12944},
     url = {https://onlinelibrary.wiley.com/doi/abs/10.1111/risa.12944},
     eprint = {https://onlinelibrary.wiley.com/doi/pdf/10.1111/risa.12944},
     abstract = {Abstract As climate change has contributed to longer fire seasons and populations living in fire-prone ecosystems increase, wildfires have begun to affect a growing number of people. As a result, interest in understanding the wildfire evacuation decision process has increased. Of particular interest is understanding why some people leave early, some choose to stay and defend their homes, and others wait to assess conditions before making a final decision. Individuals who tend to wait and see are of particular concern given the dangers of late evacuation. To understand what factors might influence different decisions, we surveyed homeowners in three areas in the United States that recently experienced a wildfire. The Protective Action Decision Model was used to identify a suite of factors previously identified as potentially relevant to evacuation decisions. Our results indicate that different beliefs about the efficacy of a particular response or action (evacuating or staying to defend), differences in risk attitudes, and emphasis on different cues to act (e.g., official warnings, environmental cues) are key factors underlying different responses. Further, latent class analysis indicates there are two general classes of individuals: those inclined to evacuate and those inclined to stay, and that a substantial portion of each class falls into the wait and see category.},
@@ -122,12 +121,19 @@ @software{forest_fire
   url = {https://github.com/sahandrez/gym_forestfire},
 }
 
-@InProceedings{Julian2018,
-    author = {Kyle D. Julian and Mykel J. Kochenderfer},
-    booktitle = gnc,
-    title = {Autonomous distributed wildfire surveillance using deep reinforcement learning},
-    year = {2018},
-    doi = {10.2514/6.2018-1589},
+@article{Julian2019,
+  title = {Distributed Wildfire Surveillance with Autonomous Aircraft Using Deep Reinforcement Learning},
+  volume = {42},
+  ISSN = {1533-3884},
+  url = {http://dx.doi.org/10.2514/1.G004106},
+  DOI = {10.2514/1.g004106},
+  number = {8},
+  journal = {Journal of Guidance,  Control,  and Dynamics},
+  publisher = {American Institute of Aeronautics and Astronautics (AIAA)},
+  author = {Julian,  Kyle D. and Kochenderfer,  Mykel J.},
+  year = {2019},
+  month = aug,
+  pages = {1768–1778}
 }
 
 @book{kochenderfer2022algorithms,

paper/paper.md

@@ -73,7 +73,7 @@ Whether or not to evacuate. If evacuating, the agent must choose a specific popu
 
 ## Modeling the Spread of Wildfires
 
-Finally, our stochastic wildfire model is based on prior work [@Julian2018]:
+Finally, our stochastic wildfire model is based on prior work [@Julian2019]:
 
 ### Fuel