Loading...
Loading...
Offline RL trajectories, game data, robot demonstrations, RLHF, multi-agent interaction
10,010 datasets
Ka To Cheung's research dataset supports a multi-study program investigating the relationship between spatial ability and science achievement in children and adolescents. The data includes cross-sectional, longitudinal, and randomized controlled trial components, identifying extrinsic-dynamic spatial ability as a key predictor of science performance. Findings support a mutualistic model of cognitive-academic co-development and demonstrate that targeted spatial training can improve both spatial skills and science achievement.
Global reference values to support the Electronic Work Reporting process. The dataset is provided by the Social Security Administration and was last updated on April 3, 2026.
P2SAMAPA published a dataset titled 'P2 Etf Dqn Engine Dataset' on the Hugging Face platform. The dataset is likely related to applying Deep Q-Networks (DQN) to Exchange-Traded Fund (ETF) trading strategies. Its specific contents, scale, and features require verification after download as metadata is minimal.
ARLBench is a performance dataset for hyperparameter optimization in reinforcement learning, compiled by autorl-org. The data originates from several thousand experimental runs conducted to identify meaningful HPO test settings. The dataset was last updated on March 19, 2026.
De-identified participant data from a randomized controlled trial investigating the effects of unsupervised campus HIIT on sedentary male students. The dataset includes pre-test and post-test measurements for physical fitness, sleep quality, and body composition. Author Di TANG published the supporting data on figshare in March 2026.
Underway measurements of ocean carbon dioxide fugacity (fCO2), sea surface salinity, and sea surface temperature collected by the IMOS Ship of Opportunity program. Data is presented in delayed mode from vessels including RV Aurora Australis, RV Investigator, RV L'Astrolabe, and RV Tangaroa, covering the Southern Ocean, Australian shelf waters, and oceans adjacent to New Zealand. The project samples regions predicted to be sensitive to climate change and critical for ocean CO2 uptake.
IMOS collects underway meteorological and oceanographic observations from scientific and Antarctic resupply voyages in oceans adjacent to Australia. The data product includes quality-controlled bulk air-sea fluxes and input observations from vessels like the RV Investigator and Aurora Australis. Observations of wind, temperature, humidity, pressure, precipitation, and radiation are processed using the COARE Bulk Flux algorithm and delivered daily.
945,000 English tweets provide a large corpus of real-world conversations between consumers and support agents on Twitter. The dataset, created by Prady06 and last updated in March 2026, is designed to drive innovation in Natural Language Processing by matching contemporary support language.
axiom-llama-bin-cache is a dataset hosted on Kaggle. Its title and description suggest it contains cache or support data related to draft models, potentially for the Llama large language model family. The raw description indicates it supports draft models, but no further metadata on size, structure, or origin is available.
Supplementary materials provide a list of excluded studies and supporting data for a scoping review on digital treatment planning and smile simulation in dentistry. The dataset is a 19.3 KB XLSX file uploaded by author Anita Bayati. It was last updated in March 2026.
A subset of the TAGS database contains histories of known-age leopard seals from observations across the Southern Ocean, primarily at Macquarie Island. The data likely includes individual measurements such as weight, length, girth, and survival records collected between 1957 and 1999. This work was completed as part of ASAC project 90 and a snapshot was stored by the Australian Antarctic Data Centre in June 2018.
1855-1993 monthly and annual mean seawater temperature, salinity, and density measurements from 26 tide gauge sites along the U.S. coast. The dataset was compiled by NOAA's National Ocean Service from primary tide gauge records, with digitization and quality control performed by students at the Florida Institute of Technology. It captures low-frequency changes in coastal ocean conditions throughout the 20th century.
1975 to 1978 data from the Geos-3 satellite altimeter, containing over 5 million sea surface height measurements. The dataset was produced by NOAA/NODC/Laboratory for Satellite Altimetry and includes supporting geophysical information. Measurements are compressed to a 1-per-second rate using a trim mean filter.
Lusaka, Zambia provides the geographic scope for this pediatric neuroscience dataset. It contains tabular data on predictors of saccadic reaction time gathered via eye tracking. The dataset is stored in an XLS file with a size of 9.5 KB.
Zambian data from a pediatric study in Lusaka examines characteristics of young children and their association with saccadic reaction time. The dataset contains tabular data in an Excel file under a CC BY 4.0 license. It was authored by Jacqueline M. Lauer and last updated in March 2026.
Statistical analysis results from a Friedman test and Bonferroni-adjusted Mann–Whitney U post-hoc comparisons across social support types. The dataset, authored by Emre Vuraloğlu, is a 5.5 KB Excel file containing the output of these non-parametric statistical procedures.
A checkpoint file for a Vanilla Deep Q-Network (DQN) model, published on Kaggle. The specific contents, such as the model architecture, training parameters, and performance metrics, are not detailed in the provided metadata. The dataset likely contains saved model weights or training state for a reinforcement learning agent.
Two NSF-funded research cruises collected this biological dataset in the Southern Drake Passage and Scotia Sea. Measurements include pigment concentrations, particulate organic matter, macronutrients, and microbial productivity. The data capture conditions during the Southern Hemisphere late summer of 2004 and winter of 2006.
Agent components for a reinforcement learning model, including state space, action space, and reward function definitions. The dataset was authored by Di Wu and uploaded to figshare in March 2026. Its scope is limited, with a file size of 5.5 KB.
Di Wu's dataset contains hyperparameters used for a reinforcement learning agent. The data is a 5.5 KB Excel file published on figshare in March 2026. It supports research into energy demand estimation and sustainable grid management.