Tool manipulation is vital for facilitating robots to complete challenging task goals. It requires reasoning about the desired effect of the task and thus properly grasping and manipulating the tool to achieve the task. Task-agnostic grasping optimizes for grasp robustness while ignoring crucial task-specific constrain…
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,738 papers · 148 categories
Trend · papers per month
48 results for “simulated task”
Learning robot tasks or controllers using deep reinforcement learning has been proven effective in simulations. Learning in simulation has several advantages. For example, one can fully control the simulated environment, including halting motions while performing computations. Another advantage when robots are involved…
User Simulators are one of the major tools that enable offline training of task-oriented dialogue systems. For this task the Agenda-Based User Simulator (ABUS) is often used. The ABUS is based on hand-crafted rules and its output is in semantic form. Issues arise from both properties such as limited diversity and the i…
Unified framework for simulation-based inference learns a single model for multiple tasks.
problem Simulation-based inference for multiple tasks with limited model retraining.
method Unified flow-matching generative model with query-aware masking distribution.
result Competitive performance on various inference tasks and real-world problems.
DIVA generates diverse tasks for complex simulators, enabling adaptive agent training.
problem Lack of diverse training data for complex, open-ended simulators.
method Evolutionary approach using domain randomization and procedural generation.
result Successfully trains adaptive agent behavior in complex simulators.
Iterative Model-Based Reinforcement Learning Using Simulations in the Differentiable Neural Computercs.LG
We propose a lifelong learning architecture, the Neural Computer Agent (NCA), where a Reinforcement Learning agent is paired with a predictive model of the environment learned by a Differentiable Neural Computer (DNC). The agent and DNC model are trained in conjunction iteratively. The agent improves its policy in simu…
Simulation-to-real transfer is an important strategy for making reinforcement learning practical with real robots. Successful sim-to-real transfer systems have difficulty producing policies which generalize across tasks, despite training for thousands of hours equivalent real robot time. To address this shortcoming, we…
This paper introduces NPR, a technique to improve Bayesian inference for multi-modal, high-dimensional simulations.
problem Challenges in Bayesian inference for multi-modal, high-dimensional simulations.
method Introduces Neural Posterior Regularization (NPR) to enforce exploration of input parameter space.
result Empirically validated that NPR significantly improves performance on various simulation tasks.
Real-life control tasks involve matters of various substances---rigid or soft bodies, liquid, gas---each with distinct physical behaviors. This poses challenges to traditional rigid-body physics engines. Particle-based simulators have been developed to model the dynamics of these complex scenes; however, relying on app…
Deep reinforcement learning has recently shown many impressive successes. However, one major obstacle towards applying such methods to real-world problems is their lack of data-efficiency. To this end, we propose the Bottleneck Simulator: a model-based reinforcement learning method which combines a learned, factorized …
We present a novel solution to the problem of simulation-to-real transfer, which builds on recent advances in robot skill decomposition. Rather than focusing on minimizing the simulation-reality gap, we learn a set of diverse policies that are parameterized in a way that makes them easily reusable. This diversity and p…
Simulation of an Elevator Group Control Using Generative Adversarial Networks and Related AI Toolsstat.ML
Generative Adversarial Networks simulate elevator group control without extensive data.
problem Lack of historical real-world data for system testing.
method Used GANs to generate simulation data for a multi-car elevator system.
result GANs can be used as substitutes for expensive simulation runs, but fine-tuning is needed.
Simformer uses transformer models to perform flexible Bayesian inference.
problem Current simulation-based inference methods are inflexible and require fixed priors.
method Trains a probabilistic diffusion model with transformer architectures.
result Outperforms state-of-the-art methods on various benchmarks.
Through many recent successes in simulation, model-free reinforcement learning has emerged as a promising approach to solving continuous control robotic tasks. The research community is now able to reproduce, analyze and build quickly on these results due to open source implementations of learning algorithms and simula…
The paper develops a faster surrogate model for simulators using hybrid methods.
problem The need for faster validation of automotive technologies using simulators.
method Testing classical methods and building hybrid models combining them.
result A hybrid surrogate model outperforms classical methods in multivariate time series prediction.
LLMs can simulate human investment attitudes based on personality traits.
problem Investigating how LLMs mimic human investment behaviors.
method Simulated investment task using LLM personas with specific Big Five personality profiles.
result LLMs can produce meaningful behavioural differences in investment tasks that align with human traits.
This paper tackles non-identifiability in financial market simulations using multivariate time series data.
problem Non-identifiability issue in social simulation models, leading to indistinguishable simulated time series data.
method Proposes a maximization-based aggregation function to form a new calibration objective function using multiple time series features.
result Significant improvements in alleviating non-identifiability and achieving higher simulation fidelity.
We present a method for fast training of vision based control policies on real robots. The key idea behind our method is to perform multi-task Reinforcement Learning with auxiliary tasks that differ not only in the reward to be optimized but also in the state-space in which they operate. In particular, we allow auxilia…
Inertial confinement fusion (ICF) experiments are designed using computer simulations that are approximations of reality, and therefore must be calibrated to accurately predict experimental observations. In this work, we propose a novel nonlinear technique for calibrating from simulations to experiments, or from low fi…
TACTO: A Fast, Flexible, and Open-source Simulator for High-Resolution Vision-based Tactile Sensorscs.RO
TACTO simulates high-resolution touch sensing for robotics.
problem Accurate simulation of touch sensing in robotics.
method Fast, flexible, open-source simulator for vision-based tactile sensors.
result Demonstrated TACTO's effectiveness in grasping stability prediction and marble manipulation control.
Multi-Task Averagingstat.ML
We present a multi-task learning approach to jointly estimate the means of multiple independent data sets. The proposed multi-task averaging (MTA) algorithm results in a convex combination of the single-task maximum likelihood estimates. We derive the optimal minimum risk estimator and the minimax estimator, and show t…
Mastering robotic manipulation skills through reinforcement learning (RL) typically requires the design of shaped reward functions. Recent developments in this area have demonstrated that using sparse rewards, i.e. rewarding the agent only when the task has been successfully completed, can lead to better policies. Howe…
Proposes an MTL method with clustering to improve regression accuracy.
problem Improving regression accuracy by sharing information among related tasks.
method Centroid parameter for clustering tasks, separating regression and clustering parameters.
result Improves estimation and prediction accuracy for regression coefficient vectors.
Large GNNs trained with Graph Parallelism improve atomic simulation accuracy.
problem Training memory-intensive GNNs for high-order atomic interactions.
method Graph Parallelism to distribute graphs across GPUs.
result Graph-parallelized models achieve state-of-the-art results on catalyst datasets.
MINIMALIST maximizes mutual information for likelihood estimation from simulated data.
problem Learning model parameters from likelihood functions that cannot be computed.
method Maximizes mutual information between simulated data and model parameters using neural networks.
result Different methods aiming at the same optimal energy form can be directly benchmarked.
Meta-reinforcement learning algorithms can enable robots to acquire new skills much more quickly, by leveraging prior experience to learn how to learn. However, much of the current research on meta-reinforcement learning focuses on task distributions that are very narrow. For example, a commonly used meta-reinforcement…
Recent advances in artificial intelligence have been driven by the presence of increasingly realistic and complex simulated environments. However, many of the existing environments provide either unrealistic visuals, inaccurate physics, low task complexity, restricted agent perspective, or a limited capacity for intera…
Enhances PCE surrogates using transfer learning for expensive simulations.
problem Over-sampling in PCE for expensive forward models.
method Transfer learning from similar tasks to a new task with limited training data.
result Improves scalability and accuracy of PCE surrogates.
SA-FDR uses simulated annealing for feature selection in high-dimensional data.
problem Feature selection in high-dimensional datasets with high predictive accuracy.
method Simulated Annealing for combinatorial optimisation of feature subsets.
result SA-FDR selects more compact feature subsets with high predictive accuracy.
New method improves ABI for sequential data, reducing forgetting and improving accuracy.
problem Performance degradation of ABI under model misspecification and distribution shifts.
method Decouples simulation-based pre-training from unsupervised SC fine-tuning, using memory buffer and elastic weight consolidation.
result Significant mitigation of forgetting and improved posterior estimates compared to standard simulation-based training.
Transferring knowledge across tasks to improve data-efficiency is one of the open key challenges in the field of global black-box optimization. Readily available algorithms are typically designed to be universal optimizers and, therefore, often suboptimal for specific tasks. We propose a novel transfer learning method …
Task-adaptive clustering improves few-shot learning with unlabeled data.
problem Handling unseen tasks with limited labeled data.
method Task-conditioned clustering in a new projection space.
result State-of-the-art semi-supervised few-shot classification performance.
Paper uses learned summary statistics for Bayesian inference with difficult likelihood functions.
problem Difficult to obtain exact likelihood function for observation data and simulation model.
method Simulation-based inference with learned summary statistics, using Cressie-Read discrepancy criterion.
result Effective inference performed over selected sample sets of observation data.
TRADES generates realistic market simulations for financial modeling.
problem Generating realistic and responsive market simulations for financial tasks.
method TRADES uses a transformer-based denoising diffusion probabilistic engine to generate time series order flows conditioned on market state.
result TRADES improves market simulation metrics by 3.27-3.48 over state-of-the-art (SoTA) methods.
We present DiffTaichi, a new differentiable programming language tailored for building high-performance differentiable physical simulators. Based on an imperative programming language, DiffTaichi generates gradients of simulation steps using source code transformations that preserve arithmetic intensity and parallelism…
Identifying context-specific entity networks from aggregated data is an important task, arising often in bioinformatics and neuroimaging. Computationally, this task can be formulated as jointly estimating multiple different, but related, sparse Undirected Graphical Models (UGM) from aggregated samples across several co…
Reliable and effective multi-task learning is a prerequisite for the development of robotic agents that can quickly learn to accomplish related, everyday tasks. However, in the reinforcement learning domain, multi-task learning has not exhibited the same level of success as in other domains, such as computer vision. In…
Residual neural networks improve collision prediction in planetary simulations.
problem Accurate prediction of planetary collisions in N-body simulations.
method Residual neural networks trained on collision data.
result Residual neural networks outperform existing methods in prediction accuracy and generalization.
Efficient Bayesian decision-making with intractable likelihoods.
problem Bayesian decision-making under intractable likelihoods.
method Learning surrogate models and using simulation-based inference and Bayesian optimization.
result Optimal actions can be learned with fewer simulations than posterior inference.
Computer simulation provides an automatic and safe way for training robotic control policies to achieve complex tasks such as locomotion. However, a policy trained in simulation usually does not transfer directly to the real hardware due to the differences between the two environments. Transfer learning using domain ra…
Online field experiments are the gold-standard way of evaluating changes to real-world interactive machine learning systems. Yet our ability to explore complex, multi-dimensional policy spaces - such as those found in recommendation and ranking problems - is often constrained by the limited number of experiments that c…
Assisted by neural networks, reinforcement learning agents have been able to solve increasingly complex tasks over the last years. The simulation environment in which the agents interact is an essential component in any reinforcement learning problem. The environment simulates the dynamics of the agents' world and henc…
MT-HAL learns features and task associations for multiple tasks with a shared sparse structure.
problem Learning features and task associations for multiple tasks with shared structure.
method Fully nonparametric approach that learns features, samples, and task associations with a shared sparse structure.
result MT-HAL achieves a powerful convergence rate and outperforms other methods across various simulation settings.
Deep multitask learning boosts performance by sharing learned structure across related tasks. This paper adapts ideas from deep multitask learning to the setting where only a single task is available. The method is formalized as pseudo-task augmentation, in which models are trained with multiple decoders for each task.…
Selective inference improves multi-task neuroimaging analysis.
problem Improving predictive performance and modeling accuracy in neuroimaging studies.
method Proposes a framework for selective inference to jointly identify relevant covariates and conduct valid inference in a sparsity-inducing model.
result Selective inference yields tighter confidence intervals and more accurate signal recovery than single-task methods.
Vision and learning have made significant progress that could improve robotics policies for complex tasks and environments. Learning deep neural networks for image understanding, however, requires large amounts of domain-specific visual data. While collecting such data from real robots is possible, such an approach lim…
Adapts deep learning models trained on simulated images for use with real images.
problem Difficulty in training deep neural networks on large amounts of experimental data.
method Adversarial domain adaptation method to mitigate domain shift between simulated and experimental image data.
result Adversarial domain adaptation successfully mitigates domain shift and improves numerical observer performance.
Unified detector calibration and simulation using MLE from generative models.
problem Combining detector calibration and simulation using traditional methods.
method Maximum likelihood estimation from conditional generative models.
result Prior-independent and non-Gaussian resolutions possible.