We focus on the problem of how we can enable a robot to collaborate seamlessly with a human partner, specifically in scenarios like collaborative manufacturing where prexisting data is sparse. Much prior work in human-robot collaboration uses observational models of humans (i.e. models that treat the robot purely as an observer) to choose the robot's behavior, but such models do not account for the influence the robot has on the human's actions, which may lead to inefficient interactions. We instead formulate the problem of optimally choosing a collaborative robot's behavior based on a conditional model of the human that depends on the robot's future behavior. First, we propose a novel model-based formulation of conditional behavior prediction that allows the robot to infer the human's intentions based on its future plan in data-sparse environments. We then show how to utilize a conditional model for proactive goal selection and path generation around human collaborators. Finally, we use our proposed proactive controller in a collaborative task with real users to show that it can improve users' interactions with a robot collaborator quantitatively and qualitatively.

Reinforcement learning from human feedback (RLHF) is a technique for training AI systems to align with human goals. RLHF has emerged as the central method used to finetune state-of-the-art large language models (LLMs). Despite this popularity, there has been relatively little public work systematizing its flaws. In this paper, we (1) survey open problems and fundamental limitations of RLHF and related methods; (2) overview techniques to understand, improve, and complement RLHF in practice; and (3) propose auditing and disclosure standards to improve societal oversight of RLHF systems. Our work emphasizes the limitations of RLHF and highlights the importance of a multi-faceted approach to the development of safer AI systems.

To act in the world, robots rely on a representation of salient task aspects: for example, to carry a cup of coffee, a robot must consider movement efficiency and cup orientation in its behaviour. However, if we want robots to act for and with people, their representations must not be just functional but also reflective of what humans care about, i.e. their representations must be aligned with humans'. In this survey, we pose that current reward and imitation learning approaches suffer from representation misalignment, where the robot's learned representation does not capture the human's representation. We suggest that because humans will be the ultimate evaluator of robot performance in the world, it is critical that we explicitly focus our efforts on aligning learned task representations with humans, in addition to learning the downstream task. We advocate that current representation learning approaches in robotics should be studied from the perspective of how well they accomplish the objective of representation alignment. To do so, we mathematically define the problem, identify its key desiderata, and situate current robot learning methods within this formalism. We conclude the survey by suggesting future directions for exploring open challenges.

Humans have internal models of robots (like their physical capabilities), the world (like what will happen next), and their tasks (like a preferred goal). However, human internal models are not always perfect: for example, it is easy to underestimate a robot's inertia. Nevertheless, these models change and improve over time as humans gather more experience. Interestingly, robot actions influence what this experience is, and therefore influence how people's internal models change. In this work we take a step towards enabling robots to understand the influence they have, leverage it to better assist people, and help human models more quickly align with reality. Our key idea is to model the human's learning as a nonlinear dynamical system which evolves the human's internal model given new observations. We formulate a novel optimization problem to infer the human's learning dynamics from demonstrations that naturally exhibit human learning. We then formalize how robots can influence human learning by embedding the human's learning dynamics model into the robot planning problem. Although our formulations provide concrete problem statements, they are intractable to solve in full generality. We contribute an approximation that sacrifices the complexity of the human internal models we can represent, but enables robots to learn the nonlinear dynamics of these internal models. We evaluate our inference and planning methods in a suite of simulated environments and an in-person user study, where a 7DOF robotic arm teaches participants to be better teleoperators. While influencing human learning remains an open problem, our results demonstrate that this influence is possible and can be helpful in real human-robot interaction.

The long-standing challenge of building effective classification models for small and imbalanced datasets has seen little improvement since the creation of the Synthetic Minority Over-sampling Technique (SMOTE) over 20 years ago. Though GAN based models seem promising, there has been a lack of purpose built architectures for solving the aforementioned problem, as most previous studies focus on applying already existing models. This paper proposes a unique, performance-oriented, data-generating strategy that utilizes a new architecture, coined draGAN, to generate both minority and majority samples. The samples are generated with the objective of optimizing the classification model's performance, rather than similarity to the real data. We benchmark our approach against state-of-the-art methods from the SMOTE family and competitive GAN based approaches on 94 tabular datasets with varying degrees of imbalance and linearity. Empirically we show the superiority of draGAN, but also highlight some of its shortcomings. All code is available on: https://github.com/LeonGuertler/draGAN.

Shared autonomy refers to approaches for enabling an autonomous agent to collaborate with a human with the aim of improving human performance. However, besides improving performance, it may often also be beneficial that the agent concurrently accounts for preserving the user's experience or satisfaction of collaboration. In order to address this additional goal, we examine approaches for improving the user experience by constraining the number of interventions by the autonomous agent. We propose two model-free reinforcement learning methods that can account for both hard and soft constraints on the number of interventions. We show that not only does our method outperform the existing baseline, but also eliminates the need to manually tune a black-box hyperparameter for controlling the level of assistance. We also provide an in-depth analysis of intervention scenarios in order to further illuminate system understanding.

We study the design of autonomous agents that are capable of deceiving outside observers about their intentions while carrying out tasks in stochastic, complex environments. By modeling the agent's behavior as a Markov decision process, we consider a setting where the agent aims to reach one of multiple potential goals while deceiving outside observers about its true goal. We propose a novel approach to model observer predictions based on the principle of maximum entropy and to efficiently generate deceptive strategies via linear programming. The proposed approach enables the agent to exhibit a variety of tunable deceptive behaviors while ensuring the satisfaction of probabilistic constraints on the behavior. We evaluate the performance of the proposed approach via comparative user studies and present a case study on the streets of Manhattan, New York, using real travel time distributions.

Existing approaches for generating human-aware agent behaviors have considered different measures of interpretability in isolation. Further, these measures have been studied under differing assumptions, thus precluding the possibility of designing a single framework that captures these measures under the same assumptions. In this paper, we present a unifying Bayesian framework that models a human observer's evolving beliefs about an agent and thereby define the problem of Generalized Human-Aware Planning. We will show that the definitions of interpretability measures like explicability, legibility and predictability from the prior literature fall out as special cases of our general framework. Through this framework, we also bring a previously ignored fact to light that the human-robot interactions are in effect open-world problems, particularly as a result of modeling the human's beliefs over the agent. Since the human may not only hold beliefs unknown to the agent but may also form new hypotheses about the agent when presented with novel or unexpected behaviors.

Specifying reward functions for robots that operate in environments without a natural reward signal can be challenging, and incorrectly specified rewards can incentivise degenerate or dangerous behavior. A promising alternative to manually specifying reward functions is to enable robots to infer them from human feedback, like demonstrations or corrections. To interpret this feedback, robots treat as approximately optimal a choice the person makes from a choice set, like the set of possible trajectories they could have demonstrated or possible corrections they could have made. In this work, we introduce the idea that the choice set itself might be difficult to specify, and analyze choice set misspecification: what happens as the robot makes incorrect assumptions about the set of choices from which the human selects their feedback. We propose a classification of different kinds of choice set misspecification, and show that these different classes lead to meaningful differences in the inferred reward and resulting performance. While we would normally expect misspecification to hurt, we find that certain kinds of misspecification are neither helpful nor harmful (in expectation). However, in other situations, misspecification can be extremely harmful, leading the robot to believe the opposite of what it should believe. We hope our results will allow for better prediction and response to the effects of misspecification in real-world reward inference.

Existing approaches for the design of interpretable agent behavior consider different measures of interpretability in isolation. In this paper we posit that, in the design and deployment of human-aware agents in the real world, notions of interpretability are just some among many considerations; and the techniques developed in isolation lack two key properties to be useful when considered together: they need to be able to 1) deal with their mutually competing properties; and 2) an open world where the human is not just there to interpret behavior in one specific form. To this end, we consider three well-known instances of interpretable behavior studied in existing literature -- namely, explicability, legibility, and predictability -- and propose a revised model where all these behaviors can be meaningfully modeled together. We will highlight interesting consequences of this unified model and motivate, through results of a user study, why this revision is necessary.