PCI combines perception and control using Bayesian inference with object-based representations.
problem Separate perception and control in reinforcement learning.
method Joint Perception and Control as Inference (PCI) framework with Object-based Perception Control (OPC).
result OPC achieves good perceptual grouping quality and outperforms baselines in accumulated rewards.
CLIC learns to control objects in non-rewarding environments by imitating a mentor.
problem Learning in non-rewarding, multi-object environments with non-observable intentions.
method Curriculum Learning and Imitation (CLIC) for Control.
result CLIC effectively learns and imitates a mentor's interactions to control objects.
The paper solves stochastic control problems with implicit objectives, finding equilibrium strategies.
problem Stochastic control problems with implicitly defined objectives leading to time-inconsistency.
method Closed-loop equilibrium solutions in a controlled diffusion framework, providing sufficient and necessary conditions.
result Explicit characterization of equilibrium portfolio strategies in terms of ordinary differential equations.
Agents use object-oriented reasoning to solve problems more quickly.
problem Finding solutions to problems more efficiently.
method Hierarchical controller directs low-level agent to simulate alternate states of the world.
result Achieves similar reward levels as non-hierarchical agents but with better data efficiency.
Proposes a method to learn dynamic models for systems with variable number of objects.
problem Efficiently modeling systems with a variable number of objects.
method Uses graph neural networks and block-wise linear transition matrices to learn compositional Koopman operators.
result The method adapts to new environments and produces better control signals.
The paper tackles non-cumulative objectives in reinforcement learning and proposes modifications to existing algorithms.
problem Optimizing objectives that are not naturally expressed as summations of rewards in various fields.
method The paper modifies the Bellman optimality equation to handle non-cumulative objectives by replacing summation with a generalized operation.
result The modified Bellman updates can converge to the globally optimal solution under certain conditions.
Paper identifies objective mismatch in MBRL, affecting control task performance.
problem Objective mismatch in MBRL framework affects control task performance.
method Proposes re-weighting dynamics model training to mitigate mismatch.
result Likelihood of one-step ahead predictions is not always correlated with control performance.
A framework disentangles controllable objects from visual signals for improved RL.
problem Improving sample efficiency and game performance in vision-based RL.
method Action-conditioned video prediction to disentangle controllable objects.
result Improved sample efficiency and game performance in Atari games.
Controller seeks informative system observations to predict nonlinear dynamics.
problem Predicting nonlinear dynamics with uncertain parameters.
method Expected free energy minimization for balancing goal state and informative observations.
result Controller improves performance in uncertain parameter scenarios.
New model learning objective improves continuous control tasks.
problem Challenges in solving continuous control tasks using model-based reinforcement learning.
method Derived a novel value-aware model learning objective and identified and addressed stale value estimates issue.
result Value-aware objectives can be successfully deployed in solving continuous control tasks without tuning hyper-parameters.
DPA aligns LLMs with multi-objective rewards for diverse user preferences.
problem Fine-grained control over LLMs for diverse user needs.
method Integrates multi-objective reward modeling and directional preference control.
result DPA offers better performance trade-offs and intuitive user control over LLM generation.
Decomposes game environments into controllable and non-controllable parts for faster learning.
problem Difficulties in end-to-end learning for complex environments.
method Modularization of the environment into controllable and non-controllable parts, using supervised and reinforcement learning.
result Human-level performance achieved in 10-15 minutes of game time with proper decomposition.
New method improves reliability of object detection models.
problem Lack of reliability in neural networks for safety-critical applications.
method Sequential Conformal Risk Control (SeqCRC) for object detection.
result Valid statistical guarantees for any dataset size without prior knowledge.
A new multi-objective RL framework improves intrinsic exploration performance.
problem Sub-optimal exploration performance due to ad-hoc handling of intrinsic exploration.
method A multi-objective RL framework where both exploration and exploitation are optimized as separate objectives.
result EMU-Q method outperforms classic and other intrinsic RL methods on benchmarks.
A new reinforcement learning method uses mutual information to encourage agents to control their environment.
problem Learning from internal drives instead of external rewards.
method Formulate an intrinsic objective as mutual information between goal states and controllable states, derive a surrogate objective for efficient optimization.
result Demonstrated the efficacy of the approach in robotic tasks.
Paper tackles stochastic control with mean and higher-order moments, finding Nash equilibria.
problem Time-inconsistent stochastic control problems with mean and higher-order moments.
method Developed closed-loop and open-loop Nash equilibrium controls using PDEs and maximum principles.
result Identical closed-loop and open-loop Nash equilibria controls, independent of state value and random path.
Gradient descent on LSE objectives implicitly performs EM, leading to collapse without volume control.
problem Gradient collapse in autoencoders without volume control.
method Introduced a single-layer encoder with an LSE objective and InfoMax regularization for volume control.
result Gradient--responsibility identity holds exactly; LSE alone collapses; variance prevents dead components; decorrelation prevents redundancy.
This paper proposes a method to safely adjust exploration in RL to satisfy constraints.
problem Unsafe exploration in reinforcement learning violates constraints on controlled object states.
method Automatic adjustment of exploration inputs and variance-covariance matrix for safety.
result The method guarantees satisfaction of joint chance constraints with specified probability.
Develops scalable differentiable physics for complex object interactions.
problem Limited scalability of existing differentiable physics solvers.
method Adopting meshes for arbitrary geometry, localized collision handling, and accelerated implicit differentiation.
result Significantly reduces memory and computation requirements compared to particle-based methods.
A framework for data-driven decision-making in infectious disease control.
problem Optimizing trade-offs between public health and economic impacts.
method Multi-objective model-based reinforcement learning.
result Pareto-optimal policies minimizing long-term costs.
MILLION framework optimizes portfolio risk and return efficiently.
problem Optimizing risk and return in AI for FinTech portfolio management.
method Two phases: return maximization with auxiliary objectives and risk control with portfolio interpolation and improvement.
result Framework achieves fine-grained risk control and improved return rates.
New control methods for systems with adversarial perturbations.
problem Control systems with adversarial noise.
method Online convex optimization and convex relaxations.
result Low regret policies against adversarial perturbations.
New method reduces variance in complex probabilistic model optimization.
problem High variance in stochastic optimisation of complex models.
method Use recognition network to approximate optimal control variate for each mini-batch.
result Sub-optimal variance reduction is improved with new approach.
New model optimizes portfolios over multiple periods using predictive control.
problem Optimizing multi-period portfolios with risk and variance objectives.
method Model Predictive Control with Mean-Variance and Risk Parity.
result 30x faster and more robust solutions compared to single period models.
This work ensures policy gradient methods converge to global optima for certain control problems.
problem Non-convex optimization challenges in policy gradient methods for complex control problems.
method Identifies structural properties ensuring non-convex objective functions have no suboptimal stationary points.
result Policy gradient methods converge to global optima under certain conditions, satisfying a Polyak-Lojasiewicz condition.
New objective reduces bias and variance in reinforcement learning derivatives.
problem Estimating derivatives in reinforcement learning with unknown dynamics.
method Derives an objective function compatible with any advantage estimators, allowing trade-off between bias and variance.
result Demonstrates effectiveness in both theoretical and practical settings.
Develops neural network framework for risk-reward optimization problems.
problem Multi-period risk-reward optimization with constrained policies.
method Neural network framework with two coupled feedforward networks, parametrizing two-step policies.
result Empirical optimum converges to true optimal value as network capacity and training size increase.
Method solves learning problem with hierarchical control objectives.
problem Learning high-dimensional nonlinear functions with model validation accuracy.
method Successive approximation method in functional spaces for hierarchical optimal control.
result Nested algorithm for solving optimal control problem.
Combines Gaussian processes and polynomial chaos for stochastic control.
problem Uncertainties in dynamic models lead to performance issues in predictive control.
method Combines Gaussian processes with polynomial chaos expansions to estimate probability distributions of nonlinear functions.
result Demonstrates accurate approximation and closed-loop performance in stochastic nonlinear model predictive control.
AutoRL automates reward tuning for RL, improving performance on complex tasks.
problem Suboptimal policies when using direct rewards in RL.
method Evolutionary search over RL agents to find optimal rewards.
result Improves RL performance on complex tasks.
Optimizes data power control in cell-free networks for better spectral efficiency.
problem Maximizing overall spectral efficiency in cell-free networks with multi-objective optimisation.
method Applied scalable multi-objective Bayesian optimisation to solve convergence-time limitations.
result Improved radio resource management in cell-free networks.
Proposes using logical specifications for multi-objective reinforcement learning to improve agent behavior.
problem Difficulties in controlling reinforcement learning agents and the need for better generalization.
method Uses propositional logic to specify the importance of multiple objectives, encoding these specifications using a recurrent neural network.
result MORL agents parameterized by logical specifications can generalize to novel combinations of objectives and achieve comparable performance.
This paper investigates the control of an ML component within the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) devoted to black-box optimization. The known CMA-ES weakness is its sample complexity, the number of evaluations of the objective function needed to approximate the global optimum. This weakness is…
Self-guidance controls image generation by extracting properties from diffusion model representations.
problem Generating images from text descriptions is challenging due to the complexity of visual details.
method Self-guidance uses internal representations of diffusion models to control image generation.
result Properties like object shape, location, and appearance can be extracted and used to steer image generation.
Pareto Testing optimizes model performance under multiple constraints.
problem Optimizing machine learning models with multiple conflicting objectives.
method Two-stage process combining optimization and statistical testing.
result Models can be configured to satisfy multiple statistical guarantees and objectives.
Two clustering algorithms optimize edge controller placement in wireless networks.
problem Optimizing edge controller placement in wireless edge networks.
method Deterministic annealing based clustering algorithms ECP-LL and ECP-LB.
result The algorithms achieve better balance between synchronization and delay costs.
Variational inference improves training of generative flow networks.
problem Training generative flow networks efficiently and accurately.
method Define variational objectives in terms of KL divergences and optimize convex combinations.
result Variational inference methods can reduce the variance of gradients in training generative flow networks.
We consider continuous-time stochastic optimal control problems featuring Conditional Value-at-Risk (CVaR) in the objective. The major difficulty in these problems arises from time-inconsistency, which prevents us from directly using dynamic programming. To resolve this challenge, we convert to an equivalent bilevel op…
Learning data representations that are transferable and are fair with respect to certain protected attributes is crucial to reducing unfair decisions while preserving the utility of the data. We propose an information-theoretically motivated objective for learning maximally expressive representations subject to fairnes…
TossingBot learns to throw objects accurately with residual physics.
problem Learning to throw arbitrary objects accurately and quickly.
method End-to-end formulation that learns control parameters from visual observations.
result TossingBot achieves 600+ grasps per hour with 85% throwing accuracy.
Derives time-averaged active inference from control principles.
problem Finite-horizon or discounted-surprise problems in active inference.
method Derives infinite-horizon, average-surprise active inference from optimal control principles.
result Unified objective functional for sensorimotor control.
Paper develops a neural-fuzzy controller for GPS-intelligent buoys.
problem Optimally track dynamically positioned marine buoys with unknown parameters.
method Dynamic system modeling using neural-fuzzy networks with backstepping technique.
result The controller minimizes position errors and adjusts buoy positions accurately.
Framework for robust RL in continuous control with model misspecification.
problem Model misspecification in reinforcement learning for continuous control.
method Integrates robustness into MPO algorithm through worst-case expected return objective and entropy regularization.
result Robust and soft-robust policies outperform non-robust policies in various domains.
Differentiable CEM enables end-to-end learning of non-convex optimization problems.
problem Non-convex optimization of continuous, parameterized objective functions.
method Introducing a differentiable variant of the cross-entropy method (CEM).
result Differentiation of CEM output with respect to parameters enables end-to-end learning.
Paper uses RL to synthesize control policies for LTL objectives in uncertain environments.
problem Control policies for uncertain, probabilistic environments with temporal logic specifications.
method Model-free RL algorithm, LTL to LDBA translation, synchronous reward function, asymptotic satisfaction probability.
result RL algorithm maximizes satisfaction probability of LTL objectives in PL-MDPs.
We propose a method to compute optimal control paths for autonomous vehicles deployed for the purpose of inferring a velocity field. In addition to being advected by the flow, the vehicles are able to effect a fixed relative speed with arbitrary control over direction. It is this direction that is used as the basis for…
RP1 uses active learning to improve world model in fewest samples.
problem Improving sample efficiency in MBRL for continuous control tasks.
method RP1 views MBRL as active learning, using a hybrid objective function and principled termination.
result Statistically significant gains over existing approaches on continuous control tasks.
The paper uses conformal prediction to detect railway signals with confidence.
problem Deploying deep learning models in certified systems requires accurate uncertainty estimates.
method The paper uses conformal prediction and risk control to detect railway signals.
result The conformal prediction framework provides reliable and trustworthy uncertainty estimates for model performance.