Experimental data are often costly to obtain, which makes it difficult to calibrate complex models. For many models an experimental design that produces the best calibration given a limited experimental budget is not obvious. This paper introduces a deep reinforcement learning (RL) algorithm for design of experiments that maximizes the information gain measured by Kullback–Leibler divergence obtained via the Kalman filter (KF). This combination enables experimental design for rapid online experiments where manual trial-and-error is not feasible in the high-dimensional parametric design space. We formulate possible configurations of experiments as a decision tree and a Markov decision process, where a finite choice of actions is available at each incremental step. Once an action is taken, a variety of measurements are used to update the state of the experiment. This new data leads to a Bayesian update of the parameters by the KF, which is used to enhance the state representation. In contrast to the Nash–Sutcliffe efficiency index, which requires additional sampling to test hypotheses for forward predictions, the KF can lower the cost of experiments by directly estimating the values of new data acquired through additional actions. In this work our applications focus on mechanical testing of materials. Numerical experiments with complex, history-dependent models are used to verify the implementation and benchmark the performance of the RL-designed experiments.
more »
« less
Choice Type Impacts Human Reinforcement Learning
Abstract In reinforcement learning (RL) experiments, participants learn to make rewarding choices in response to different stimuli; RL models use outcomes to estimate stimulus–response values that change incrementally. RL models consider any response type indiscriminately, ranging from more concretely defined motor choices (pressing a key with the index finger), to more general choices that can be executed in a number of ways (selecting dinner at the restaurant). However, does the learning process vary as a function of the choice type? In Experiment 1, we show that it does: Participants were slower and less accurate in learning correct choices of a general format compared with learning more concrete motor actions. Using computational modeling, we show that two mechanisms contribute to this. First, there was evidence of irrelevant credit assignment: The values of motor actions interfered with the values of other choice dimensions, resulting in more incorrect choices when the correct response was not defined by a single motor action; second, information integration for relevant general choices was slower. In Experiment 2, we replicated and further extended the findings from Experiment 1 by showing that slowed learning was attributable to weaker working memory use, rather than slowed RL. In both experiments, we ruled out the explanation that the difference in performance between two condition types was driven by difficulty/different levels of complexity. We conclude that defining a more abstract choice space used by multiple learning systems for credit assignment recruits executive resources, limiting how much such processes then contribute to fast learning.
more »
« less
- Award ID(s):
- 2020844
- NSF-PAR ID:
- 10446606
- Date Published:
- Journal Name:
- Journal of Cognitive Neuroscience
- Volume:
- 35
- Issue:
- 2
- ISSN:
- 0898-929X
- Page Range / eLocation ID:
- 314 to 330
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Abstract How does the similarity between stimuli affect our ability to learn appropriate response associations for them? In typical laboratory experiments learning is investigated under somewhat ideal circumstances, where stimuli are easily discriminable. This is not representative of most real-life learning, where overlapping “stimuli” can result in different “rewards” and may be learned simultaneously (e.g., you may learn over repeated interactions that a specific dog is friendly, but that a very similar looking one isn’t). With two experiments, we test how humans learn in three stimulus conditions: one “best case” condition in which stimuli have idealized and highly discriminable visual and semantic representations, and two in which stimuli have overlapping representations, making them less discriminable. We find that, unsurprisingly, decreasing stimuli discriminability decreases performance. We develop computational models to test different hypotheses about how reinforcement learning (RL) and working memory (WM) processes are affected by different stimulus conditions. Our results replicate earlier studies demonstrating the importance of both processes to capture behavior. However, our results extend previous studies by demonstrating that RL, and not WM, is affected by stimulus distinctness: people learn slower and have higher across-stimulus value confusion at decision when stimuli are more similar to each other. These results illustrate strong effects of stimulus type on learning and demonstrate the importance of considering parallel contributions of different cognitive processes when studying behavior.
-
more » « less
Abstract Reinforcement learning research has pursued a persistent question: Does reward feedback prompt inferences that transcend simple associations? Reversal learning data suggest an affirmative answer: When the positive stimulus (S+) becomes the negative stimulus (S−), trained humans rapidly switch to choosing the former S−. The operations supporting such inferences remain ambiguous. Do participants identify transitions between stimulus‐specific contexts (i.e., A+B− and A−B+), or deduce values by learning the abstract contingency structure? Across two experiments, we probed humans’ use of abstract rules to infer the values of unchosen alternatives. In Experiment 1, 37 participants attempted a task that originally demonstrated monkeys’ difficulty with this form of inference. We presented modified discrimination problems in which the initially chosen stimulus (abstract inference group) or unchosen stimulus (control group) was replaced with a novel stimulus of identical status on Trial 2. In the abstract inference condition, accurate performance can be achieved by applying the consistent contingency structure (but not memory of stimulus‐specific reward associations) to infer to the unchosen stimulus’ value. The abstract inference group learned to make accurate choices, but only after committing substantially more errors than were observed among control participants—suggesting that unchosen value inferences are infrequently drawn in standard discrimination scenarios. In Experiment 2, 17 participants completed abstract inference problems that had been modified to be suitable for
fMRI investigations. Behavioral results both corroborated the Experiment 1 trends and further revealed marked individual differences in explicit awareness of the novel stimulus values. -
more » « less
Background: Short-term forecasts of infectious disease burden can contribute to situational awareness and aid capacity planning. Based on best practice in other fields and recent insights in infectious disease epidemiology, one can maximise the predictive performance of such forecasts if multiple models are combined into an ensemble. Here, we report on the performance of ensembles in predicting COVID-19 cases and deaths across Europe between 08 March 2021 and 07 March 2022.
Methods: We used open-source tools to develop a public European COVID-19 Forecast Hub. We invited groups globally to contribute weekly forecasts for COVID-19 cases and deaths reported by a standardised source for 32 countries over the next 1–4 weeks. Teams submitted forecasts from March 2021 using standardised quantiles of the predictive distribution. Each week we created an ensemble forecast, where each predictive quantile was calculated as the equally-weighted average (initially the mean and then from 26th July the median) of all individual models’ predictive quantiles. We measured the performance of each model using the relative Weighted Interval Score (WIS), comparing models’ forecast accuracy relative to all other models. We retrospectively explored alternative methods for ensemble forecasts, including weighted averages based on models’ past predictive performance.
Results: Over 52 weeks, we collected forecasts from 48 unique models. We evaluated 29 models’ forecast scores in comparison to the ensemble model. We found a weekly ensemble had a consistently strong performance across countries over time. Across all horizons and locations, the ensemble performed better on relative WIS than 83% of participating models’ forecasts of incident cases (with a total N=886 predictions from 23 unique models), and 91% of participating models’ forecasts of deaths (N=763 predictions from 20 models). Across a 1–4 week time horizon, ensemble performance declined with longer forecast periods when forecasting cases, but remained stable over 4 weeks for incident death forecasts. In every forecast across 32 countries, the ensemble outperformed most contributing models when forecasting either cases or deaths, frequently outperforming all of its individual component models. Among several choices of ensemble methods we found that the most influential and best choice was to use a median average of models instead of using the mean, regardless of methods of weighting component forecast models.
Conclusions: Our results support the use of combining forecasts from individual models into an ensemble in order to improve predictive performance across epidemiological targets and populations during infectious disease epidemics. Our findings further suggest that median ensemble methods yield better predictive performance more than ones based on means. Our findings also highlight that forecast consumers should place more weight on incident death forecasts than incident case forecasts at forecast horizons greater than 2 weeks.
Funding: AA, BH, BL, LWa, MMa, PP, SV funded by National Institutes of Health (NIH) Grant 1R01GM109718, NSF BIG DATA Grant IIS-1633028, NSF Grant No.: OAC-1916805, NSF Expeditions in Computing Grant CCF-1918656, CCF-1917819, NSF RAPID CNS-2028004, NSF RAPID OAC-2027541, US Centers for Disease Control and Prevention 75D30119C05935, a grant from Google, University of Virginia Strategic Investment Fund award number SIF160, Defense Threat Reduction Agency (DTRA) under Contract No. HDTRA1-19-D-0007, and respectively Virginia Dept of Health Grant VDH-21-501-0141, VDH-21-501-0143, VDH-21-501-0147, VDH-21-501-0145, VDH-21-501-0146, VDH-21-501-0142, VDH-21-501-0148. AF, AMa, GL funded by SMIGE - Modelli statistici inferenziali per governare l'epidemia, FISR 2020-Covid-19 I Fase, FISR2020IP-00156, Codice Progetto: PRJ-0695. AM, BK, FD, FR, JK, JN, JZ, KN, MG, MR, MS, RB funded by Ministry of Science and Higher Education of Poland with grant 28/WFSN/2021 to the University of Warsaw. BRe, CPe, JLAz funded by Ministerio de Sanidad/ISCIII. BT, PG funded by PERISCOPE European H2020 project, contract number 101016233. CP, DL, EA, MC, SA funded by European Commission - Directorate-General for Communications Networks, Content and Technology through the contract LC-01485746, and Ministerio de Ciencia, Innovacion y Universidades and FEDER, with the project PGC2018-095456-B-I00. DE., MGu funded by Spanish Ministry of Health / REACT-UE (FEDER). DO, GF, IMi, LC funded by Laboratory Directed Research and Development program of Los Alamos National Laboratory (LANL) under project number 20200700ER. DS, ELR, GG, NGR, NW, YW funded by National Institutes of General Medical Sciences (R35GM119582; the content is solely the responsibility of the authors and does not necessarily represent the official views of NIGMS or the National Institutes of Health). FB, FP funded by InPresa, Lombardy Region, Italy. HG, KS funded by European Centre for Disease Prevention and Control. IV funded by Agencia de Qualitat i Avaluacio Sanitaries de Catalunya (AQuAS) through contract 2021-021OE. JDe, SMo, VP funded by Netzwerk Universitatsmedizin (NUM) project egePan (01KX2021). JPB, SH, TH funded by Federal Ministry of Education and Research (BMBF; grant 05M18SIA). KH, MSc, YKh funded by Project SaxoCOV, funded by the German Free State of Saxony. Presentation of data, model results and simulations also funded by the NFDI4Health Task Force COVID-19 (
https://www.nfdi4health.de/task-force-covid-19-2 ) within the framework of a DFG-project (LO-342/17-1). LP, VE funded by Mathematical and Statistical modelling project (MUNI/A/1615/2020), Online platform for real-time monitoring, analysis and management of epidemic situations (MUNI/11/02202001/2020); VE also supported by RECETOX research infrastructure (Ministry of Education, Youth and Sports of the Czech Republic: LM2018121), the CETOCOEN EXCELLENCE (CZ.02.1.01/0.0/0.0/17-043/0009632), RECETOX RI project (CZ.02.1.01/0.0/0.0/16-013/0001761). NIB funded by Health Protection Research Unit (grant code NIHR200908). SAb, SF funded by Wellcome Trust (210758/Z/18/Z). -
null (Ed.)If our choices make us who we are, then what does that mean when these choices are made in the human-machine interface? Developing a clear understanding of how human decision making is influenced by automated systems in the environment is critical because, as human-machine interfaces and assistive robotics become even more ubiquitous in everyday life, many daily decisions will be an emergent result of the interactions between the human and the machine – not stemming solely from the human. For example, choices can be influenced by the relative locations and motor costs of the response options, as well as by the timing of the response prompts. In drift diffusion model simulations of response-prompt timing manipulations, we find that it is only relatively equibiased choices that will be successfully influenced by this kind of perturbation. However, with drift diffusion model simulations of motor cost manipulations, we find that even relatively biased choices can still show some influence of the perturbation. We report the results of a two-alternative forced-choice experiment with a computer mouse modified to have a subtle velocity bias in a pre-determined direction for each trial, inducing an increased motor cost to move the cursor away from the pre-designated target direction. With queries that have each been normed in advance to be equibiased in people’s preferences, the participant will often begin their mouse movement before their cognitive choice has been finalized, and the directional bias in the mouse velocity exerts a small but significant influence on their final choice. With queries that are not equibiased, a similar influence is observed. By exploring the synergies that are developed between humans and machines and tracking their temporal dynamics, this work aims to provide insight into our evolving decisions.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1162/jocn_a_01947
