This study focuses on embodied agents that can follow natural language instructions to complete complex tasks in a visually-perceived environment. Existing methods rely on a large amount of (instruction, gold trajectory) pairs to learn a good policy. The high data cost and poor sample efficiency prevents the development of versatile agents that are capable of many tasks and can learn new tasks quickly. In this work, we propose a novel method, LLM-Planner, that harnesses the power of large language models (LLMs) such as GPT-3 to do few-shot planning for embodied agents. We further propose a simple but effective way to enhance LLMs with physical grounding to generate plans that are grounded in the current environment. Experiments on the ALFRED dataset show that our method can achieve very competitive few-shot performance, even outperforming several recent baselines that are trained using the full training data despite using less than 0.5% of paired training data. Existing methods can barely complete any task successfully under the same few-shot setting. Our work opens the door for developing versatile and sample-efficient embodied agents that can quickly learn many tasks.
Recent interest in integrated sensing and communications has led to the design of novel signal processing techniques to recover information from an overlaid radar-communications signal. Here, we focus on a spectral coexistence scenario, wherein the channels and transmit signals of both radar and communications systems are unknown to the common receiver. In this dual-blind deconvolution (DBD) problem, the receiver admits a multi-carrier wireless communications signal that is overlaid with the radar signal reflected off multiple targets. The communications and radar channels are represented by continuous-valued range-times or delays corresponding to multiple transmission paths and targets, respectively. Prior works addressed recovery of unknown channels and signals in this ill-posed DBD problem through atomic norm minimization but contingent on individual minimum separation conditions for radar and communications channels. In this paper, we provide an optimal joint separation condition using extremal functions from the Beurling-Selberg interpolation theory. Thereafter, we formulate DBD as a low-rank modified Hankel matrix retrieval and solve it via nuclear norm minimization. We estimate the unknown target and communications parameters from the recovered low-rank matrix using multiple signal classification (MUSIC) method. We show that the joint separation condition also guarantees that the underlying Vandermonde matrix for MUSIC is well-conditioned. Numerical experiments validate our theoretical findings.
Consider a target being tracked by a cognitive radar network. If the target can intercept some radar network emissions, how can it detect coordination among the radars? By 'coordination' we mean that the radar emissions satisfy Pareto optimality with respect to multi-objective optimization over each radar's utility. This paper provides a novel multi-objective inverse reinforcement learning approach which allows for both detection of such Pareto optimal ('coordinating') behavior and subsequent reconstruction of each radar's utility function, given a finite dataset of radar network emissions. The method for accomplishing this is derived from the micro-economic setting of Revealed Preferences, and also applies to more general problems of inverse detection and learning of multi-objective optimizing systems.
Distributed machine learning enables scalability and computational offloading, but requires significant levels of communication. Consequently, communication efficiency in distributed learning settings is an important consideration, especially when the communications are wireless and battery-driven devices are employed. In this paper we develop a censoring-based heavy ball (CHB) method for distributed learning in a server-worker architecture. Each worker self-censors unless its local gradient is sufficiently different from the previously transmitted one. The significant practical advantages of the HB method for learning problems are well known, but the question of reducing communications has not been addressed. CHB takes advantage of the HB smoothing to eliminate reporting small changes, and provably achieves a linear convergence rate equivalent to that of the classical HB method for smooth and strongly convex objective functions. The convergence guarantee of CHB is theoretically justified for both convex and nonconvex cases. In addition we prove that, under some conditions, at least half of all communications can be eliminated without any impact on convergence rate. Extensive numerical results validate the communication efficiency of CHB on both synthetic and real datasets, for convex, nonconvex, and nondifferentiable cases. Given a target accuracy, CHB can significantly reduce the number of communications compared to existing algorithms, achieving the same accuracy without slowing down the optimization process.
In mathematical psychology, recent models for human decision-making use Quantum Decision Theory to capture important human-centric features such as order effects and violation of the sure-thing principle (total probability law). We construct and analyze a human-sensor system where a quickest detector aims to detect a change in an underlying state by observing human decisions that are influenced by the state. Apart from providing an analytical framework for such human-sensor systems, we also analyze the structure of the quickest detection policy. We show that the quickest detection policy has a single threshold and the optimal cost incurred is lower bounded by that of the classical quickest detector. This indicates that intermediate human decisions strictly hinder detection performance. We also analyze the sensitivity of the quickest detection cost with respect to the quantum decision parameters of the human decision maker, revealing that the performance is robust to inaccurate knowledge of the decision-making process. Numerical results are provided which suggest that observing the decisions of more rational decision makers will improve the quickest detection performance. Finally, we illustrate a numerical implementation of this quickest detector in the context of the Prisoner's Dilemma problem, in which it has been observed that Quantum Decision Theory can uniquely model empirically tested violations of the sure-thing principle.
In federated learning (FL), the objective of collaboratively learning a global model through aggregation of model updates across devices tends to oppose the goal of personalization via local information. In this work, we calibrate this tradeoff in a quantitative manner through a multi-criterion optimization-based framework, which we cast as a constrained program: the objective for a device is its local objective, which it seeks to minimize while satisfying nonlinear constraints that quantify the proximity between the local and the global model. By considering the Lagrangian relaxation of this problem, we develop an algorithm that allows each node to minimize its local component of Lagrangian through queries to a first-order gradient oracle. Then, the server executes Lagrange multiplier ascent steps followed by a Lagrange multiplier-weighted averaging step. We call this instantiation of the primal-dual method Federated Learning Beyond Consensus ($\texttt{FedBC}$). Theoretically, we establish that $\texttt{FedBC}$ converges to a first-order stationary point at rates that matches the state of the art, up to an additional error term that depends on the tolerance parameter that arises due to the proximity constraints. Overall, the analysis is a novel characterization of primal-dual methods applied to non-convex saddle point problems with nonlinear constraints. Finally, we demonstrate that $\texttt{FedBC}$ balances the global and local model test accuracy metrics across a suite of datasets (Synthetic, MNIST, CIFAR-10, Shakespeare), achieving competitive performance with the state of the art.
In federated learning (FL), the objective of collaboratively learning a global model through aggregation of model updates across devices tends to oppose the goal of personalization via local information. In this work, we calibrate this tradeoff in a quantitative manner through a multi-criterion optimization-based framework, which we cast as a constrained program: the objective for a device is its local objective, which it seeks to minimize while satisfying nonlinear constraints that quantify the proximity between the local and the global model. By considering the Lagrangian relaxation of this problem, we develop an algorithm that allows each node to minimize its local component of Lagrangian through queries to a first-order gradient oracle. Then, the server executes Lagrange multiplier ascent steps followed by a Lagrange multiplier-weighted averaging step. We call this instantiation of the primal-dual method Federated Learning Beyond Consensus ($\texttt{FedBC}$). Theoretically, we establish that $\texttt{FedBC}$ converges to a first-order stationary point at rates that matches the state of the art, up to an additional error term that depends on the tolerance parameter that arises due to the proximity constraints. Overall, the analysis is a novel characterization of primal-dual methods applied to non-convex saddle point problems with nonlinear constraints. Finally, we demonstrate that $\texttt{FedBC}$ balances the global and local model test accuracy metrics across a suite of datasets (Synthetic, MNIST, CIFAR-10, Shakespeare), achieving competitive performance with the state of the art.
We consider a joint multiple-antenna radar-communications system in a co-existence scenario. Contrary to conventional applications, wherein at least the radar waveform and communications channel are known or estimated \textit{a priori}, we investigate the case when the channels and transmit signals of both systems are unknown. In radar applications, this problem arises in multistatic or passive systems, where transmit signal is not known. Similarly, highly dynamic vehicular or mobile communications may render prior estimates of wireless channel unhelpful. In particular, the radar signal reflected-off multiple targets is overlaid with the multi-carrier communications signal. In order to extract the unknown continuous-valued target parameters (range, Doppler velocity, and direction-of-arrival) and communications messages, we formulate the problem as a sparse dual-blind deconvolution and solve it using atomic norm minimization. Numerical experiments validate our proposed approach and show that precise estimation of continuous-valued channel parameters, radar waveform, and communications messages is possible up to scaling ambiguities.
We consider a team of autonomous agents that navigate in an adversarial environment and aim to achieve a task by allocating their resources over a set of target locations. The adversaries in the environment observe the autonomous team's behavior to infer their objective and counter-allocate their own resources to the target locations. In this setting, we develop strategies for controlling the density of the autonomous team so that they can deceive the adversaries regarding their objective while achieving the desired final resource allocation. We first develop a prediction algorithm, based on the principle of maximum entropy, to express the team's behavior expected by the adversaries. Then, by measuring the deceptiveness via Kullback-Leibler divergence, we develop convex optimization-based planning algorithms that deceives adversaries by either exaggerating the behavior towards a decoy allocation strategy or creating ambiguity regarding the final allocation strategy. Finally, we illustrate the performance of the proposed algorithms through numerical simulations.
In this work, we propose a novel ${\bf K}$ernelized ${\bf S}$tein Discrepancy-based Posterior Sampling for ${\bf RL}$ algorithm (named $\texttt{KSRL}$) which extends model-based RL based upon posterior sampling (PSRL) in several ways: we (i) relax the need for any smoothness or Gaussian assumptions, allowing for complex mixture models; (ii) ensure it is applicable to large-scale training by incorporating a compression step such that the posterior consists of a \emph{Bayesian coreset} of only statistically significant past state-action pairs; and (iii) develop a novel regret analysis of PSRL based upon integral probability metrics, which, under a smoothness condition on the constructed posterior, can be evaluated in closed form as the kernelized Stein discrepancy (KSD). Consequently, we are able to improve the $\mathcal{O}(H^{3/2}d\sqrt{T})$ {regret} of PSRL to $\mathcal{O}(H^{3/2}\sqrt{T})$, where $d$ is the input dimension, $H$ is the episode length, and $T$ is the total number of episodes experienced, alleviating a linear dependence on $d$ . Moreover, we theoretically establish a trade-off between regret rate with posterior representational complexity via introducing a compression budget parameter $\epsilon$ based on KSD, and establish a lower bound on the required complexity for consistency of the model. Experimentally, we observe that this approach is competitive with several state of the art RL methodologies, with substantive improvements in computation time. Experimentally, we observe that this approach is competitive with several state of the art RL methodologies, and can achieve up-to $50\%$ reduction in wall clock time in some continuous control environments.