Expectation Alignment: Handling Reward Misspecification in the Presence of Expectation Mismatch

• URL: http://arxiv.org/abs/2404.08791v2
• Date: Thu, 31 Oct 2024 02:34:00 GMT
• Title: Expectation Alignment: Handling Reward Misspecification in the Presence of Expectation Mismatch
• Authors: Malek Mechergui, Sarath Sreedharan,
• Abstract summary: We use the theory of mind, i.e., the human user's beliefs about the AI agent, as a basis to develop a formal explanatory framework. We propose a new interactive algorithm that uses the specified reward to infer potential user expectations.
• Score: 19.03141646688652
• License:
• Abstract: Detecting and handling misspecified objectives, such as reward functions, has been widely recognized as one of the central challenges within the domain of Artificial Intelligence (AI) safety research. However, even with the recognition of the importance of this problem, we are unaware of any works that attempt to provide a clear definition for what constitutes (a) misspecified objectives and (b) successfully resolving such misspecifications. In this work, we use the theory of mind, i.e., the human user's beliefs about the AI agent, as a basis to develop a formal explanatory framework called Expectation Alignment (EAL) to understand the objective misspecification and its causes. Our EAL framework not only acts as an explanatory framework for existing works but also provides us with concrete insights into the limitations of existing methods to handle reward misspecification and novel solution strategies. We use these insights to propose a new interactive algorithm that uses the specified reward to infer potential user expectations about the system behavior. We show how one can efficiently implement this algorithm by mapping the inference problem into linear programs. We evaluate our method on a set of standard Markov Decision Process (MDP) benchmarks.

Related papers

• Know Where You're Uncertain When Planning with Multimodal Foundation Models: A Formal Framework [54.40508478482667]
  We present a comprehensive framework to disentangle, quantify, and mitigate uncertainty in perception and plan generation. We propose methods tailored to the unique properties of perception and decision-making. We show that our uncertainty disentanglement framework reduces variability by up to 40% and enhances task success rates by 5% compared to baselines.
  arXiv  Detail & Related papers  (2024-11-03T17:32:00Z)
• Towards Measuring Goal-Directedness in AI Systems [0.0]
  A key prerequisite for AI systems pursuing unintended goals is whether they will behave in a coherent and goal-directed manner. We propose a new family of definitions of the goal-directedness of a policy that analyze whether it is well-modeled as near-optimal for many reward functions. Our contribution is a definition of goal-directedness that is simpler and more easily computable in order to approach the question of whether AI systems could pursue dangerous goals.
  arXiv  Detail & Related papers  (2024-10-07T01:34:42Z)
• Interpretable Concept-Based Memory Reasoning [12.562474638728194]
  Concept-based Memory Reasoner (CMR) is a novel CBM designed to provide a human-understandable and provably-verifiable task prediction process. CMR achieves better accuracy-interpretability trade-offs to state-of-the-art CBMs, discovers logic rules consistent with ground truths, allows for rule interventions, and allows pre-deployment verification.
  arXiv  Detail & Related papers  (2024-07-22T10:32:48Z)
• Goal Recognition via Linear Programming [14.129476759815251]
  Research on Goal Recognition as Planning encompasses reasoning about the model of a planning task, the observations, and the goals using planning techniques. In this article, we design novel recognition approaches that rely on the Operator-Counting framework. We show how the new IP/LP constraints can improve the recognition of goals under both partial and noisy observability.
  arXiv  Detail & Related papers  (2024-04-11T17:34:35Z)
• Clarify When Necessary: Resolving Ambiguity Through Interaction with LMs [58.620269228776294]
  We propose a task-agnostic framework for resolving ambiguity by asking users clarifying questions. We evaluate systems across three NLP applications: question answering, machine translation and natural language inference. We find that intent-sim is robust, demonstrating improvements across a wide range of NLP tasks and LMs.
  arXiv  Detail & Related papers  (2023-11-16T00:18:50Z)
• Safe Explicable Planning [3.3869539907606603]
  We propose Safe Explicable Planning (SEP) to support the specification of a safety bound. Our approach generalizes the consideration of multiple objectives stemming from multiple models. We provide formal proofs that validate the desired theoretical properties of these methods.
  arXiv  Detail & Related papers  (2023-04-04T21:49:02Z)
• Goal Alignment: A Human-Aware Account of Value Alignment Problem [16.660807368368758]
  Value alignment problems arise in scenarios where the specified objectives of an AI agent don't match the true underlying objective of its users. A foundational cause for misalignment is the inherent asymmetry in human expectations about the agent's behavior and the behavior generated by the agent for the specified objective. We propose a novel formulation for the value alignment problem, named goal alignment that focuses on a few central challenges related to value alignment.
  arXiv  Detail & Related papers  (2023-02-02T01:18:57Z)
• Discrete Factorial Representations as an Abstraction for Goal Conditioned Reinforcement Learning [99.38163119531745]
  We show that applying a discretizing bottleneck can improve performance in goal-conditioned RL setups. We experimentally prove the expected return on out-of-distribution goals, while still allowing for specifying goals with expressive structure.
  arXiv  Detail & Related papers  (2022-11-01T03:31:43Z)
• MURAL: Meta-Learning Uncertainty-Aware Rewards for Outcome-Driven Reinforcement Learning [65.52675802289775]
  We show that an uncertainty aware classifier can solve challenging reinforcement learning problems. We propose a novel method for computing the normalized maximum likelihood (NML) distribution. We show that the resulting algorithm has a number of intriguing connections to both count-based exploration methods and prior algorithms for learning reward functions.
  arXiv  Detail & Related papers  (2021-07-15T08:19:57Z)
• Learning Uncertainty For Safety-Oriented Semantic Segmentation In Autonomous Driving [77.39239190539871]
  We show how uncertainty estimation can be leveraged to enable safety critical image segmentation in autonomous driving. We introduce a new uncertainty measure based on disagreeing predictions as measured by a dissimilarity function. We show experimentally that our proposed approach is much less computationally intensive at inference time than competing methods.
  arXiv  Detail & Related papers  (2021-05-28T09:23:05Z)
• Outcome-Driven Reinforcement Learning via Variational Inference [95.82770132618862]
  We discuss a new perspective on reinforcement learning, recasting it as the problem of inferring actions that achieve desired outcomes, rather than a problem of maximizing rewards. To solve the resulting outcome-directed inference problem, we establish a novel variational inference formulation that allows us to derive a well-shaped reward function. We empirically demonstrate that this method eliminates the need to design reward functions and leads to effective goal-directed behaviors.
  arXiv  Detail & Related papers  (2021-04-20T18:16:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.