Pith. sign in

REVIEW 2 major objections 6 minor 59 references

To Measure or Not: A Cost-Sensitive, Selective Measuring Environment for Agricultural Management Decisions with Reinforcement Learning

T0 review · 2 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A recurrent PPO agent can simultaneously learn nitrogen fertilization and selective crop measurement in a WOFOST wheat environment, and with realistic per-feature costs it beats a fixed three-date schedule while measuring about a quarter…

desk verdict Useful environment paper with a genuine first, but the headline yield advantage over standard practice is not statistically supported. read the letter →

arxiv 2501.12823 v1 pith:YCC3GCTJ submitted 2025-01-22 cs.LG cs.AI

Hilmy Baja , Michiel Kallenberg , Ioannis N. Athanasiadis This is my paper · ORCID
classification cs.LGcs.AI
keywords reinforcementlearningcropmanagementactivefeatureacquisitionmeasurementcostsnitrogenfertilizationWOFOSTwinterwheatpartiallyobservableMDP
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Farmers rarely have cheap, complete crop-state data before every fertilizer decision, yet most RL crop-management studies assume those observations are free. This paper tries to remove that assumption by making measurement itself a decision: in a weekly WOFOST winter-wheat simulation, a recurrent PPO agent chooses both a nitrogen application rate and which of six crop features to measure, paying an explicit cost for each measurement. The paper claims that with realistic feature costs the agent measures only about a quarter of the weeks, concentrates measurements near critical development stages, and still achieves a higher median yield (7.46 t/ha) than a fixed three-date fertilizer schedule (7.30 t/ha). It also finds that higher measurement costs degrade performance, and that in cost-free settings the agent measures roughly half the steps and matches an agent given complete observations. A sympathetic reader would care because the result suggests data-driven fertilizer optimization can be made practical under realistic data-collection budgets.

What carries the argument

The load-bearing object is an active-feature-acquisition POMDP (a partially observable Markov decision process where the action space is a control action plus a measurement vector), extended so each feature has its own cost. Each week the control action is one of seven nitrogen rates, and the measurement action is a vector of binary choices for six crop features; the reward subtracts the chosen feature costs, a fixed field-deployment cost (10), and a fertilizer cost (β=2) from the weekly gain in simulated wheat yield (TWSO). Unmeasured features are presented to the policy as masked zeros, and the agent's recurrent LSTM memory must carry information across weeks. A single recurrent PPO (proximal policy optimization with LSTM actor and critic) jointly optimizes both action heads over a 47-week season in the WOFOST simulator. The mechanism is cost-driven: because DVS (development stage) and weather are always free, the agent can time its paid measurements to moments, such as flowering, where they most improve fertilizer decisions.

What would settle it

Re-run the Realistic-cost experiment with the cost vector scaled by 0.5, 1, 2, and 5, keeping everything else fixed, and compute the agent's net reward (yield value minus fertilizer, deployment, and measurement costs). If at any plausible scale the Realistic-cost policy's net reward falls below the fixed three-date schedule's net reward, the paper's claim that realistic costs still beat standard practice is falsified.

Watch

Extended reading notes

Core claim

The paper's central claim is that a single RL agent can learn a useful joint policy for applying nitrogen and for deciding when to pay for crop-feature measurements, and that doing so is necessary for realistic crop management. The authors adapt the AFA-POMDP formulation by assigning each measurable feature its own cost and appending observation masks to the input: unmeasured features appear as masked zeros, so the agent must remember past observations through its recurrent network. In the Realistic-cost scenario, the LSTM-PPO agent pays for cheap features (LAI, soil moisture) about five times a season, rarely buys the expensive lab-type measurements, and almost never measures a random distractor feature. The resulting median yield of 7.46 t/ha beats the standard-practice baseline of 7.30 t/ha and comes within about 5% of the cost-free upper bound of 7.86 t/ha. This is taken as evidence that selective, cost-aware measuring can approach complete-observation performance while cutting data collection effort.

Load-bearing premise

The whole comparison rests on the hand-assigned feature costs (5, 5, 20, 20, 25, 10), the fertilizer price ratio β=2, and the deployment cost D=10 being realistic; if actual measurement prices differ, the learned policy and the yield ranking could change.

Editorial extensions

If this is right

  • Under the paper's cost assumptions, a cost-aware RL policy needs far fewer observations than prior crop-management RL: about 12 of 47 weekly steps involved any paid measurement in the Realistic scenario, versus roughly half the steps when measurements were free.
  • Expensive features get used only when they are likely to matter: NuptakeTotal is measured about twice as often as NAVAIL despite equal cost, and the random distractor feature is measured least in every costed scenario.
  • The learned measuring policy is weather-adaptive: in the cold year (2010) the agent delays both measurements and fertilization relative to the normal year (1990), following the delayed crop development visible through free DVS observations.
  • Measurement costs change attainable yield: median yields fall monotonically from 7.86 t/ha (No-cost) to 6.63 t/ha (Exp-cost), so realistic cost assumptions are not neutral for policy evaluation.
  • Because the Realistic scenario beats the fixed three-date standard practice and the None-observed agent, the paper concludes that targeted measurement rather than either no observation or complete observation is the right operating point.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The hand-set cost vector (5 for LAI and SM, 10 for the distractor, 20 for NuptakeTotal and NAVAIL, 25 for TAGP) is the switch that decides everything; a natural extension is to run the same training loop with cost multipliers taken from actual soil-lab and sensor-service price lists and check whether the Realistic policy still beats standard practice.
  • The same measure-and-control formulation could be moved from nitrogen to other costly sensing tasks, such as irrigation scheduling with soil-moisture probes or pest scouting, where the information value of a measurement has to justify its acquisition cost.
  • The paper's noiseless non-destructive measurement assumption probably understates the value of redundant measurement: with sensor noise and lab turnaround delays, measuring the same feature twice or measuring earlier may become rational, so field trials could reveal more measuring than the simulated policy.
  • Because the agent must be retrained for new sites, a cost-aware measurement policy learned in one region may not transfer; training on multiple sites with randomized initial soil conditions would test whether the flowering-stage measuring heuristic generalizes.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 6 minor

Summary. The paper proposes an active feature acquisition formulation for crop management, extending AFA-POMDPs to include feature-specific measurement costs. An LSTM-based PPO agent acts in a WOFOST winter wheat environment with a weekly cadence, jointly deciding nitrogen fertilization amounts (0-60 kg/ha in 10 kg/ha increments) and whether to measure each of six crop features (LAI, SM, NuptakeTotal, NAVAIL, TAGP, and a random distraction feature). The environment is evaluated in four cost scenarios (No-cost, Flat-cost, Realistic, Exp-cost) plus two non-measuring RL baselines (All-observed, None-observed) and two fixed policy baselines (Standard-practice, Random-spread). The authors report that higher measurement costs hurt yield, that the Realistic-cost agent measures most often near flowering and preferentially measures cheaper features, and that the Realistic-cost agent achieves higher median yield than the standard fixed-date fertilization baseline. The paper also documents that the agent learns to ignore the random feature when measurements are costly.

Significance. If the performance claims are substantiated, the paper would provide a practically relevant demonstration that RL can simultaneously optimize fertilization and data collection, reducing measurement burden while maintaining yield. The manuscript's strengths include a carefully specified and released configurable environment (CropGym-ToMeasureOrNot), a clear training/evaluation split with 16 training years and 16 held-out years, multiple seeds, and a sensible set of baselines including a standard-practice fertilization schedule. The finding that the agent measures less often for costly, low-information features and aligns measurement times with critical crop development stages is a useful qualitative result. However, the central quantitative claim of outperforming standard practice rests on a small median difference with overlapping confidence intervals, and the cost parameters that define the 'Realistic' scenario are hand-assigned. These issues make the current evidence conditional and require targeted additional analysis before the main claims can be accepted.

major comments (2)
  1. [Experiments and Results, Table 3] The central claim that the Realistic-cost agent 'manages to achieve better performance compared to a baseline of standard practice' is not statistically supported by the reported results. The median yields are 7.46 t/ha for Realistic and 7.30 t/ha for Standard-practice, but the bootstrapped 95% confidence intervals are (6.45, 9.13) and (6.55, 8.65), respectively, showing substantial overlap. Since all scenarios are evaluated on the same seeded years and locations, a paired test (e.g., Wilcoxon signed-rank test or paired bootstrap over the 10 seeds and 16 evaluation years) is required to determine whether the observed margin is systematic or due to seed/year noise. Please report such a paired comparison for Realistic versus Standard-practice, and also for the other pairwise claims that depend on Table 3 (e.g., Flat-cost vs Realistic). If the paired test is not significant, the statement of outperformance should be softened accordingly.
  2. [Design Rationale and Assumptions / Table 1] The feature measurement costs in Table 1 (LAI=5, SM=5, NuptakeTotal=20, NAVAIL=20, TAGP=25, Random=10) and the reward parameters beta=2 and D=10 are treated as fixed ground truth in all experiments. The qualitative justification is plausible, but the entire cost-scenario comparison, including the learned measuring policy in Table 2 and the performance ordering in Table 3, is conditional on these values. Different but equally realistic cost estimates could change the learned policy and the ranking of Realistic versus Standard-practice. I ask for a sensitivity analysis that varies the cost vector (e.g., scaling the expensive features by 0.5x, 2x, and 3x) and the reward parameters (beta and D) over a credible range, and reports whether the qualitative conclusions—especially the cost scenario ordering and the comparison with standard practice—remain stable. Without this, the 'realistic' scenario is an uncalibrated assumption rather than a validated setting.
minor comments (6)
  1. [Experiments and Results, Table 2] The text says Table 2 reports 'the average number of measurement actions performed by the agent in a one-year period', but the table shows values that are averaged across years and seeds. Please clarify the aggregation explicitly (e.g., mean over 16 evaluation years and 10 seeds) and state the standard deviation or MAD consistently in the caption.
  2. [Random feature measuring policy] The sentence 'We set the Realistic cost to 10, same as Flat-cost' is ambiguous because it could be read as the entire Realistic cost vector being 10; in fact, Table 1 sets only the Random feature's Realistic cost to 10. Please rephrase to specify that the Random feature's cost is 10 in the Realistic scenario.
  3. [Adaptive policy] The term 'yearly cumulative minimum temperatures' (with values 1980.11 and 1488.36) is not standard and the units are unclear. Please define the quantity precisely, e.g., sum of daily minimum temperatures over the growing season in °C, and state what 'colder' means in this metric.
  4. [RL Environment, Eq. (1)] The cost term is written as sum from i=0 to Nm, while the text defines c as a vector of size Nm. The index should start at 1 (or the upper limit should be Nm-1) to avoid an off-by-one inconsistency.
  5. [Figure 3] The figure shows measurement actions as transparent vertical lines, but the legend does not distinguish which features were measured. Please indicate whether all measured features are pooled or only a subset; the caption should state this explicitly.
  6. [RL Environment, RL agent] The hyperparameter description says 'each with 2 hidden layers with size of 256' and 'The rest of the hyperparameters we kept same as the default.' Please specify which defaults (e.g., Stable Baselines 3 PPO defaults) and list the exact architecture and any altered hyperparameters for reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the measuring and fertilization policies are emergent outputs of RL training, not re-fitted inputs.

full rationale

The paper's derivation chain is empirical rather than definitional: it defines a POMDP/AFA-POMDP with a reward that subtracts explicit feature costs, trains an LSTM-PPO agent in the WOFOST simulator, and then reports the yields and measuring frequencies produced by the trained policies. The claimed discoveries (adaptive measurement near flowering, preferring cheaper informative features, ignoring the Random feature, later fertilization in cold years) are not encoded in the reward or observation space; they are observed behaviors of the learned policy. The cost schedule in Table 1 and the reward parameters are inputs to the optimization, not values fitted to the outputs, so no fitted parameter is relabeled as a prediction. Self-citations to Kallenberg et al. (2023) and earlier CropGym work are used for data-aggregation conventions and environment lineage, not as load-bearing evidence for the central claim, and no uniqueness theorem or ansatz is imported from those citations. The overlapping confidence intervals between the Realistic scenario and Standard-practice in Table 3 are a statistical-support concern, not a circularity concern. The paper is self-contained in the sense that its central results are generated inside its own defined environment with an external crop model and compared against explicit baselines, so no circular derivation is present.

Assumptions & free parameters 3 free parameters · 4 assumptions · 1 invented entities

The quantitative conclusions are conditional on WOFOST, the hand-set cost and reward parameters, and the noiseless-measurement simplification. No sensitivity analysis is given for the cost vector, so the ranking of cost scenarios is not robust to alternative realistic cost estimates.

free parameters (3)
  • Feature measurement cost vector = LAI=5, SM=5, NuptakeTotal=20, NAVAIL=20, TAGP=25, Random=10
    Hand-set based on qualitative sensor/lab cost descriptions; not fitted to market prices. These costs determine the trade-off between measuring and yield and drive the scenario comparisons.
  • Fertilizer-to-wheat price ratio beta = 2
    Chosen to mimic average prices (Wageningen Economic Research). It scales the nitrogen cost in the reward and affects how much fertilizer the agent applies.
  • Deployment cost D = 10
    Hand-chosen penalty for each fertilization event. It influences fertilization frequency and interacts with the measuring policy.
assumptions (4)
  • domain assumption WOFOST is a valid model of winter wheat growth and nitrogen response
    Used as the environment's transition function; all training and evaluation happen inside WOFOST, and its accuracy is assumed from prior validation.
  • domain assumption Measurements are noiseless, immediate, non-destructive, and incur repeatable costs
    Stated in the Assumptions subsection. This removes sensor noise, delays, and one-time sensor investments, which could materially change optimal measuring policies.
  • ad hoc to paper The Table 1 costs approximate real-world measurement prices
    No market price data or calibration is provided; the 'realistic' scenario label rests on this assumption.
  • domain assumption The fixed three-date fertilization schedule is a representative standard-practice baseline
    Used as the comparison for the claim of better performance; other farmer practices could shift the comparison.
invented entities (1)
  • Random distraction feature
    purpose: Added to the observation space to test whether the agent learns to ignore irrelevant crop-state measurements.
    A synthetic normal random variable with no relation to crop growth. It is useful only as a probe inside the constructed environment; its measuring frequency is not evidence about real-world features.

how reviews work

0 comments
Cite this review

Pith. "Pith review of To Measure or Not: A Cost-Sensitive, Selective Measuring Environment for Agricultural Management Decisions with Reinforcement Learning." pith.science (2026). https://pith.science/paper/YCC3GCTJ

@misc{pith2026250112823,
  author       = {Pith},
  title        = {Pith review of: To Measure or Not: A Cost-Sensitive, Selective Measuring Environment for Agricultural Management Decisions with Reinforcement Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YCC3GCTJ}},
  note         = {Machine review of arXiv:2501.12823}
}
read the original abstract

Farmers rely on in-field observations to make well-informed crop management decisions to maximize profit and minimize adverse environmental impact. However, obtaining real-world crop state measurements is labor-intensive, time-consuming and expensive. In most cases, it is not feasible to gather crop state measurements before every decision moment. Moreover, in previous research pertaining to farm management optimization, these observations are often assumed to be readily available without any cost, which is unrealistic. Hence, enabling optimization without the need to have temporally complete crop state observations is important. An approach to that problem is to include measuring as part of decision making. As a solution, we apply reinforcement learning (RL) to recommend opportune moments to simultaneously measure crop features and apply nitrogen fertilizer. With realistic considerations, we design an RL environment with explicit crop feature measuring costs. While balancing costs, we find that an RL agent, trained with recurrent PPO, discovers adaptive measuring policies that follow critical crop development stages, with results aligned by what domain experts would consider a sensible approach. Our results highlight the importance of measuring when crop feature measurements are not readily available.

Figures

Figures reproduced from arXiv: 2501.12823 by the authors.

Figure 1
Figure 1. A schematic diagram describing the observation and action space processing, without the [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. The normalized aggregated measuring frequency calculated from all seeds and years. We show shaded bands to [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. LAI, measurement and fertilization actions for 1990 and 2010 growing year period for the Realistic sce￾nario. The y-axis (left) shows LAI development and sec￾ondary y-axis (right) shows bars depicting applied fertilizer actions in [kg/ha]. Measurement actions are shown with transparent vertical lines. it chose to measure three times in January, which is generally the emergence period for winter wheat, signifying the… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

59 extracted references · 51 canonical work pages

  1. [1]

    A.; Kumar, U.; Ansar, M.; Shah, G

    Ahmed, M.; Ahmad, S.; Raza, M. A.; Kumar, U.; Ansar, M.; Shah, G. A.; Parsons, D.; Hoogenboom, G.; Palosuo, T.; and Seidel, S. 2020. Models calibration and evaluation. Systems modeling, 151--178

    work page 2020

  2. [2]

    Bellinger, C.; Coles, R.; Crowley, M.; and Tamblyn, I. 2021 a . Active Measure Reinforcement Learning for Observation Cost Minimization. In Canadian Conference on AI

    work page 2021

  3. [3]

    Bellinger, C.; Crowley, M.; and Tamblyn, I. 2023. Dynamic observation policies in observation cost-sensitive reinforcement learning. arXiv preprint arXiv:2307.02620

    arXiv 2023

  4. [4]

    Bellinger, C.; Drozdyuk, A.; Crowley, M.; and Tamblyn, I. 2021 b . Scientific Discovery and the Cost of Measurement-- B alancing Information and Cost in Reinforcement Learning. arXiv preprint arXiv:2112.07535

    work page Pith review arXiv 2021

  5. [5]

    N.; Silva, J

    Berghuijs, H. N.; Silva, J. V.; Reidsma, P.; and de Wit , A. J. 2024. Expanding the WOFOST crop model to explore options for sustainable nitrogen management: A study for winter wheat in the Netherlands. European Journal of Agronomy, 154: 127099

    work page 2024

  6. [6]

    Bréda, N. J. J. 2003. Ground‐based measurements of leaf area index: a review of methods, instruments and current controversies . Journal of Experimental Botany, 54(392): 2403--2417

    work page 2003

  7. [7]

    Cassandra, A. R. 1998. A survey of POMDP applications. In Working notes of AAAI 1998 fall symposium on planning with partially observable Markov decision processes, volume 1724

    work page 1998

  8. [8]

    D.; Reidsma, P.; van Vliet, M

    Chukalla, A. D.; Reidsma, P.; van Vliet, M. T.; Silva, J. V.; van Ittersum, M. K.; Jomaa, S.; Rode, M.; Merbach, I.; and van Oel, P. R. 2020. Balancing indicators for sustainable intensification of crop production at field and river basin levels. Science of the Total Environment, 705

    work page 2020

Show all 59 references
  1. [9]

    Cobbe, K.; Klimov, O.; Hesse, C.; Kim, T.; and Schulman, J. 2019. Quantifying Generalization in Reinforcement Learning. arXiv:1812.02341

    2019 arXiv

  2. [10]

    Cravero, A.; Pardo, S.; Sep \'u lveda, S.; and Mu \ n oz, L. 2022. Challenges to use machine learning in agricultural big data: a systematic literature review. Agronomy, 12(3): 748

    2022

  3. [11]

    C.; and Johnson, G

    Dahnke, W. C.; and Johnson, G. V. 1990. Testing Soils for Available Nitrogen, chapter 6, 127--139. John Wiley & Sons, Ltd. ISBN 9780891188629

    1990

  4. [12]

    Day, A.; and Intalap, S. 1970. Some effects of soil moisture stress on the growth of wheat (Triticum aestivum L. em Thell.) 1. Agronomy Journal, 62(1): 27--29

    1970

  5. [13]

    de Wit, A. 2023. The P ython C rop S imulation E nvironment

    2023

  6. [14]

    De Wit, A.; Boogaard, H.; Fumagalli, D.; Janssen, S.; Knapen, R.; van Kraalingen, D.; Supit, I.; van der Wijngaart, R.; and van Diepen, K. 2019. 25 years of the WOFOST cropping systems model. Agricultural systems, 168: 154--167

    2019

  7. [15]

    Donatelli, M.; Magarey, R.; Bregaglio, S.; Willocquet, L.; Whish, J.; and Savary, S. 2017. Modelling the impacts of pests and diseases on agricultural systems. Agricultural Systems, 155: 213--224

    2017

  8. [16]

    Eck, H. V. 1988. Winter wheat response to nitrogen and irrigation. Agronomy journal, 80(6): 902--908

    1988

  9. [17]

    M.; and Gul, F

    Fahad, M.; Ahmad, I.; Rehman, M.; Waqas, M. M.; and Gul, F. 2019. Regional wheat yield estimation by integration of remotely sensed soil moisture into a crop model. Canadian Journal of Remote Sensing, 45(6): 770--781

    2019

  10. [18]

    Flach, B.; and Selten, M. 2021. Netherlands: D utch Parliament Approves Law to Reduce Nitrogen Emissions. Global Agricultural Information Network Report NL2020-0069, Foreign Agricultural Service, US Department of Agriculture

    2021

  11. [19]

    Gautron, R.; Maillard, O.-A.; Preux, P.; Corbeels, M.; and Sabbadin, R. 2022 a . Reinforcement learning for crop management support: R eview, prospects and challenges. Computers and Electronics in Agriculture, 200: 107182

    2022

  12. [20]

    J.; Preux, P.; Bigot, J.; Maillard, O.-A.; and Emukpere, D

    Gautron, R.; Padrón, E. J.; Preux, P.; Bigot, J.; Maillard, O.-A.; and Emukpere, D. 2022 b . gym- DSSAT : a crop model turned into a Reinforcement Learning environment. arXiv:2207.03270

    2022 arXiv

  13. [21]

    Goldenits, G.; Mallinger, K.; Raubitzek, S.; and Neubauer, T. 2024. Current applications and potential future directions of reinforcement learning-based Digital Twins in agriculture. Smart Agricultural Technology, 100512

    2024

  14. [22]

    Hasegawa, K.; Matsuyama, H.; Tsuzuki, H.; and Sweda, T. 2010. Improving the estimation of leaf area index by using remotely sensed NDVI with BRDF signatures. Remote Sensing of Environment, 114(3): 514--519

    2010

  15. [23]

    He, D.; Wang, E.; Wang, J.; and Robertson, M. J. 2017. Data requirement for effective calibration of process-based crop models. Agricultural and Forest Meteorology, 234: 136--148

    2017

  16. [24]

    Hirel, B.; Le Gouis, J.; Ney, B.; and Gallais, A. 2007. The challenge of improving nitrogen use efficiency in crop plants: towards a more central role for genetic variability and quantitative genetics within integrated approaches . Journal of Experimental Botany, 58(9): 2369--2387

    2007

  17. [25]

    W.; Sudduth, K

    Hummel, J. W.; Sudduth, K. A.; and Hollinger, S. E. 2001. Soil moisture and organic matter prediction of surface and subsurface soils using an NIR soil sensor. Computers and electronics in agriculture, 32(2): 149--165

    2001

  18. [26]

    G.; Overweg, H.; van Bree, R.; and Athanasiadis, I

    Kallenberg, M. G.; Overweg, H.; van Bree, R.; and Athanasiadis, I. N. 2023. Nitrogen management with reinforcement learning and crop growth models. Environmental Data Science, 2: e34

    2023

  19. [27]

    Kossen, J.; Cangea, C.; Vértes, E.; Jaegle, A.; Patraucean, V.; Ktena, I.; Tomasev, N.; and Belgrave, D. 2023. Active Acquisition for Multimodal Temporal Data: A Challenging Decision-Making Task. arXiv:2211.05039

    2023 arXiv

  20. [28]

    D.; and Jansen, N

    Krale, M.; Sim \ a o, T. D.; and Jansen, N. 2023. Act-Then-Measure: R einforcement Learning for Partially Observable Environments with Active Measuring. arXiv preprint arXiv:2303.08271

    2023 arXiv

  21. [29]

    Kuyah, S.; and Rosenstock, T. S. 2015. Optimal measurement strategies for aboveground tree biomass in agricultural landscapes. Agroforestry systems, 89: 125--133

    2015

  22. [30]

    E.; and Wang, L

    Li, B.; Jin, J.; Zhong, H.; Hopcroft, J. E.; and Wang, L. 2022. Why Robust Generalization in Deep Learning is Difficult: Perspective of Expressive Power. ArXiv, abs/2205.13863

    2022 arXiv

  23. [31]

    M.; Baedeker, T.; Braimoh, A.; Bwalya, M.; Caron, P.; Cattaneo, A.; Garrity, D.; Henry, K.; et al

    Lipper, L.; Thornton, P.; Campbell, B. M.; Baedeker, T.; Braimoh, A.; Bwalya, M.; Caron, P.; Cattaneo, A.; Garrity, D.; Henry, K.; et al. 2014. Climate-smart agriculture for food security. Nature climate change, 4(12): 1068--1072

    2014

  24. [32]

    Ma, C.; Liu, M.; Ding, F.; Li, C.; Cui, Y.; Chen, W.; and Wang, Y. 2022. Wheat growth monitoring and yield estimation based on remote sensing data assimilation into the SAFY crop growth model. Scientific Reports, 12(1): 5473

    2022

  25. [33]

    Madondo, M.; Azmat, M.; Dipietro, K.; Horesh, R.; Jacobs, M.; Bawa, A.; Srinivasan, R.; and O'Donncha, F. 2023. A SWAT -based Reinforcement Learning Framework for Crop Management. arXiv preprint arXiv:2302.04988

    2023 arXiv

  26. [34]

    M.-S.; Olesen, J

    Martín, M. M.-S.; Olesen, J. E.; and Porter, J. R. 2014. A genotype, environment and management (GxExM) analysis of adaptation in winter wheat to climate change in Denmark. Agricultural and Forest Meteorology, 187: 1--13

    2014

  27. [35]

    Martínez-Dalmau, J.; Berbel, J.; and Ordóñez-Fernández, R. 2021. Nitrogen Fertilization. A Review of the Risks Associated with the Inefficiency of Its Use and Policy Responses. Sustainability, 13(10)

    2021

  28. [36]

    A.; Fleming, S.; and Brunskill, E

    Nam, H. A.; Fleming, S.; and Brunskill, E. 2021. Reinforcement learning with state observation costs in action-contingent noiselessly observable M arkov decision processes. Advances in Neural Information Processing Systems, 34: 15650--15666

    2021

  29. [37]

    Ni, T.; Eysenbach, B.; and Salakhutdinov, R. 2021. Recurrent Model-Free RL is a Strong Baseline for Many POMDPs. CoRR, abs/2110.05038

    2021 arXiv

  30. [38]

    Oberst, M.; and Sontag, D. A. 2019. Counterfactual Off-Policy Evaluation with Gumbel-Max Structural Causal Models. CoRR, abs/1905.05824

    2019 arXiv

  31. [39]

    Overweg, H.; Berghuijs, H. N. C.; and Athanasiadis, I. N. 2021. C rop G ym: a Reinforcement Learning Environment for Crop Management. CoRR, abs/2104.04326

    2021 arXiv

  32. [40]

    Raffin, A.; Hill, A.; Gleave, A.; Kanervisto, A.; Ernestus, M.; and Dormann, N. 2021. Stable-Baselines3 : R eliable Reinforcement Learning Implementations. Journal of Machine Learning Research, 22(268): 1--8

    2021

  33. [41]

    Schmugge, T. J. 1983. Remote Sensing of Soil Moisture: Recent Advances. IEEE Transactions on Geoscience and Remote Sensing, GE-21(3): 336--344

    1983

  34. [42]

    G.; and Giller, K

    Schut, A. G.; and Giller, K. E. 2020. Soil-based, field-specific fertilizer recommendations are a pipe-dream. Geoderma, 380: 114680

    2020

  35. [43]

    J.; and Yang, E

    Shim, H.; Hwang, S. J.; and Yang, E. 2018. Joint active feature acquisition and classification with variable-size set encoding. Advances in neural information processing systems, 31

    2018

  36. [44]

    Sparks, A. H. 2018. nasapower: a NASA POWER global meteorology, surface solar energy and climatology data client for R. Journal of Open Source Software, 3(30): 1035

    2018

  37. [45]

    F.; Harrison, M

    Tao, R.; Zhao, P.; Wu, J.; Martin, N. F.; Harrison, M. T.; Ferreira, C.; Kalantari, Z.; and Hovakimyan, N. 2022. Optimizing crop management with reinforcement learning and imitation learning. arXiv preprint arXiv:2209.09991

    2022 arXiv

  38. [46]

    M.; Bir, C.; Widmar, D

    Thompson, N. M.; Bir, C.; Widmar, D. A.; and Minert, J. R. 2019. Farmer perceptions of precision agriculture technology benefits. Journal of Agricultural and Applied Economics, 51(1): 142–163

    2019

  39. [47]

    Tobin, J.; Fong, R.; Ray, A.; Schneider, J.; Zaremba, W.; and Abbeel, P. 2017. Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World. CoRR, abs/1703.06907

    2017 arXiv

  40. [48]

    M.; and Krause, A

    Turchetta, M.; Corinzia, L.; Sussex, S.; Burton, A.; Herrera, J.; Athanasiadis, I.; Buhmann, J. M.; and Krause, A. 2022. Learning long-term crop management strategies with C ycles G ym. Advances in Neural Information Processing Systems, 35: 11396--11409

    2022

  41. [49]

    Ulissi, V.; Antonucci, F.; Benincasa, P.; Farneselli, M.; Tosti, G.; Guiducci, M.; Tei, F.; Costa, C.; Pallottino, F.; Pari, L.; et al. 2011. Nitrogen concentration estimation in tomato leaves by VIS-NIR non-destructive spectroscopy. Sensors, 11(6): 6411--6424

    2011

  42. [50]

    Van der Velde, M.; and Nisini, L. 2019. Performance of the MARS -crop yield forecasting system for the E uropean U nion: A ssessing accuracy, in-season, and year-to-year improvements from 1993 to 2015. Agricultural Systems, 168: 203--212

    2019

  43. [51]

    v.; Wolf, J

    Van Diepen, C. v.; Wolf, J. v.; Van Keulen, H.; and Rappoldt, C. 1989. WOFOST : a simulation model of crop production. Soil use and management, 5(1): 16--24

    1989

  44. [52]

    Wageningen Economic Research . 2023 a . agrimatie.nl. Price development of fertilizer. Accessed: 2023-09-16

    2023

  45. [53]

    Wageningen Economic Research . 2023 b . agrimatie.nl. Price development of seeds and grains. Accessed: 2023-09-16

    2023

  46. [54]

    B.; Qin, X.; Yan, N.; Chang, S.; Zhao, Y.; Dong, Q.; Boken, V.; Plotnikov, D.; Guo, H.; Wu, F.; Zhao, H.; Deronde, B.; Tits, L.; and Loupian, E

    Wu, B.; Zhang, M.; Zeng, H.; Tian, F.; Potgieter, A. B.; Qin, X.; Yan, N.; Chang, S.; Zhao, Y.; Dong, Q.; Boken, V.; Plotnikov, D.; Guo, H.; Wu, F.; Zhao, H.; Deronde, B.; Tits, L.; and Loupian, E. 2022 a . Challenges and opportunities in remote sensing-based crop monitoring: ...

    2022

  47. [55]

    F.; and Hovakimyan, N

    Wu, J.; Tao, R.; Zhao, P.; Martin, N. F.; and Hovakimyan, N. 2022 b . Optimizing nitrogen management with deep reinforcement learning and crop simulations. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 1712--1720

    2022

  48. [56]

    J.; Zhang, C.; and Tschiatschek, S

    Yin, H.; Li, Y.; Pan, S. J.; Zhang, C.; and Tschiatschek, S. 2020. Reinforcement learning with efficient active feature acquisition. arXiv preprint arXiv:2011.00825

    2020 arXiv

  49. [57]

    A.; Schapendonk, A

    Yin, X.; Lantinga, E. A.; Schapendonk, A. H.; and Zhong, X. 2003. Some quantitative relationships between leaf area index and canopy nitrogen content and distribution. Annals of botany, 91(7): 893--903

    2003

  50. [58]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...

  51. [59]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.