Toryn Q. Klassen

Mathieu Tuli, Andrew C. Li, Pashootan Vaezipoor et al.

Text-based games present a unique class of sequential decision making problem in which agents interact with a partially observable, simulated environment via actions and observations conveyed through natural language. Such observations typically include instructions that, in a reinforcement learning (RL) setting, can directly or indirectly guide a player towards completing reward-worthy tasks. In this work, we study the ability of RL agents to follow such instructions. We conduct experiments that show that the performance of state-of-the-art text-based game agents is largely unaffected by the presence or absence of such instructions, and that these agents are typically unable to execute tasks to completion. To further study and address the task of instruction following, we equip RL agents with an internal structured representation of natural language instructions in the form of Linear Temporal Logic (LTL), a formal language that is increasingly used for temporally extended reward specification in RL. Our framework both supports and highlights the benefit of understanding the temporal semantics of instructions and in measuring progress towards achievement of such a temporally extended behaviour. Experiments with 500+ games in TextWorld demonstrate the superior performance of our approach.

Andrew C. Li, Zizhao Chen, Pashootan Vaezipoor et al.

Natural and formal languages provide an effective mechanism for humans to specify instructions and reward functions. We investigate how to generate policies via RL when reward functions are specified in a symbolic language captured by Reward Machines, an increasingly popular automaton-inspired structure. We are interested in the case where the mapping of environment state to a symbolic (here, Reward Machine) vocabulary -- commonly known as the labelling function -- is uncertain from the perspective of the agent. We formulate the problem of policy learning in Reward Machines with noisy symbolic abstractions as a special class of POMDP optimization problem, and investigate several methods to address the problem, building on existing and new techniques, the latter focused on predicting Reward Machine state, rather than on grounding of individual symbols. We analyze these methods and evaluate them experimentally under varying degrees of uncertainty in the correct interpretation of the symbolic vocabulary. We verify the strength of our approach and the limitation of existing methods via an empirical investigation on both illustrative, toy domains and partially observable, deep RL domains.

Chandler Smith, Marwa Abdulhai, Manfred Diaz et al.

Large Language Model (LLM) agents have demonstrated impressive capabilities for social interaction and are increasingly being deployed in situations where they might engage with both human and artificial agents. These interactions represent a critical frontier for LLM-based agents, yet existing evaluation methods fail to measure how well these capabilities generalize to novel social situations. In this paper, we introduce a method for evaluating the ability of LLM-based agents to cooperate in zero-shot, mixed-motive environments using Concordia, a natural language multi-agent simulation environment. Our method measures general cooperative intelligence by testing an agent's ability to identify and exploit opportunities for mutual gain across diverse partners and contexts. We present empirical results from the NeurIPS 2024 Concordia Contest, where agents were evaluated on their ability to achieve mutual gains across a suite of diverse scenarios ranging from negotiation to collective action problems. Our findings reveal significant gaps between current agent capabilities and the robust generalization required for reliable cooperation, particularly in scenarios demanding persuasion and norm enforcement.

Dillon Z. Chen, Till Hofmann, Toryn Q. Klassen et al.

Generalised planning (GP) refers to the task of synthesising programs that solve families of related planning problems. We introduce a novel, yet simple method for GP: given a set of training problems, for each problem, compute an optimal plan for each goal atom in some order, perform goal regression on the resulting plans, and lift the corresponding outputs to obtain a set of first-order $\textit{Condition} \rightarrow \textit{Actions}$ rules. The rules collectively constitute a generalised plan that can be executed as is or alternatively be used to prune the planning search space. We formalise and prove the conditions under which our method is guaranteed to learn valid generalised plans and state space pruning axioms for search. Experiments demonstrate significant improvements over state-of-the-art (generalised) planners with respect to the 3 metrics of synthesis cost, planning coverage, and solution quality on various classical and numeric planning domains.

Parand A. Alamdari, Toryn Q. Klassen, Sheila A. McIlraith

We examine one particular dimension of AI governance: how to monitor and audit AI-enabled products and services throughout the AI development lifecycle, from pre-deployment testing to post-deployment auditing. Combining principles from formal methods with SoTA machine learning, we propose techniques that enable AI-enabled product and service developers, as well as third party AI developers and evaluators, to perform offline auditing and online (runtime) monitoring of product-specific (temporally extended) behavioral constraints such as safety constraints, norms, rules and regulations with respect to black-box advanced AI systems, notably LLMs. We further provide practical techniques for predictive monitoring, such as sampling-based methods, and we introduce intervening monitors that act at runtime to preempt and potentially mitigate predicted violations. Experimental results show that by exploiting the formal syntax and semantics of Linear Temporal Logic (LTL), our proposed auditing and monitoring techniques are superior to LLM baseline methods in detecting violations of temporally extended behavioral constraints; with our approach, even small-model labelers match or exceed frontier LLM judges. Our predictive and intervening monitors significantly reduce the violation rates of LLM-based agents while largely preserving task performance. We further show through controlled experiments that LLMs' temporal reasoning shows a pronounced degradation in accuracy with increasing event distance, number of constraints, and number of propositions.

Dillon Z. Chen, Till Hofmann, Toryn Q. Klassen et al.

We tackle the challenge of building embodied AI agents that can reliably solve long-horizon planning problems. Imitation learning from demonstrations has shown itself to be effective in training robots to solve a diversity of complex tasks requiring fine motor control and manipulation over low-level (LL), continuous environments. Yet, it remains a difficult endeavour to generate long-horizon plans from imitation learning alone. In contrast, high-level (HL), symbolic abstractions facilitate efficient and interpretable long-horizon planning. We propose to combine the strengths of LL imitation learning for manipulation and control, and HL symbolic abstractions for long-horizon planning. We realise this idea via \emph{bilevel policies} of the form $(π^{\mathrm{hl}}, π^{\mathrm{ll}})$, consisting of a neural policy $π^{\mathrm{ll}}$ learned from LL demonstrations, and an HL symbolic policy $π^{\mathrm{hl}}$ that is constructed from symbolic abstractions of the LL demonstrations combined with inductive generalisation. We implement these ideas in the BISON system. Experiments on extended MetaWorld benchmarks demonstrate that BISON generalises to long horizons and problems with greater numbers of objects than those solved by VLA and end-to-end methods, and is more time and memory efficient in training and inference. Notably, when ignoring LL execution, BISON's HL policies can solve HL problems with 10,000 relevant objects in under a minute. Project page: https://dillonzchen.github.io/bison

Parand A. Alamdari, Toryn Q. Klassen, Elliot Creager et al.

Fair decision making has largely been studied with respect to a single decision. Here we investigate the notion of fairness in the context of sequential decision making where multiple stakeholders can be affected by the outcomes of decisions. We observe that fairness often depends on the history of the sequential decision-making process, and in this sense that it is inherently non-Markovian. We further observe that fairness often needs to be assessed at time points within the process, not just at the end of the process. To advance our understanding of this class of fairness problems, we explore the notion of non-Markovian fairness in the context of sequential decision making. We identify properties of non-Markovian fairness, including notions of long-term, anytime, periodic, and bounded fairness. We explore the interplay between non-Markovian fairness and memory and how memory can support construction of fair policies. Finally, we introduce the FairQCM algorithm, which can automatically augment its training data to improve sample efficiency in the synthesis of fair policies via reinforcement learning.

Parand A. Alamdari, Toryn Q. Klassen, Rodrigo Toro Icarte et al.

Pluralistic alignment is concerned with ensuring that an AI system's objectives and behaviors are in harmony with the diversity of human values and perspectives. In this paper we study the notion of pluralistic alignment in the context of agentic AI, and in particular in the context of an agent that is trying to learn a policy in a manner that is mindful of the values and perspective of others in the environment. To this end, we show how being considerate of the future wellbeing and agency of other (human) agents can promote a form of pluralistic alignment.

Giovanni Varricchione, Toryn Q. Klassen, Natasha Alechina et al.

Reward machines (RMs) are automata structures that encode (non-Markovian) reward functions for reinforcement learning (RL). RMs can reward any behaviour representable in regular languages and, when paired with RL algorithms that exploit RM structure, have been shown to significantly improve sample efficiency in many domains. In this work, we present pushdown reward machines (pdRMs), an extension of reward machines based on deterministic pushdown automata. pdRMs can recognise and reward temporally extended behaviours representable in deterministic context-free languages, making them more expressive than reward machines. We introduce two variants of pdRM-based policies, one which has access to the entire stack of the pdRM, and one which can only access the top $k$ symbols (for a given constant $k$) of the stack. We propose a procedure to check when the two kinds of policies (for a given environment, pdRM, and constant $k$) achieve the same optimal state values. We then provide theoretical results establishing the expressive power of pdRMs, and space complexity results for the proposed learning problems. Lastly, we propose an approach for off-policy RL algorithms that exploits counterfactual experiences with pdRMs. We conclude by providing experimental results showing how agents can be trained to perform tasks representable in deterministic context-free languages using pdRMs.

Toryn Q. Klassen, Parand A. Alamdari, Sheila A. McIlraith

If an AI system makes decisions over time, how should we evaluate how aligned it is with a group of stakeholders (who may have conflicting values and preferences)? In this position paper, we advocate for consideration of temporal aspects including stakeholders' changing levels of satisfaction and their possibly temporally extended preferences. We suggest how a recent approach to evaluating fairness over time could be applied to a new form of pluralistic alignment: temporal pluralism, where the AI system reflects different stakeholders' values at different times.

Andrew C. Li, Toryn Q. Klassen, Andrew Wang et al.

Grounding language in perception and action is a key challenge when building situated agents that can interact with humans, or other agents, via language. In the past, addressing this challenge has required manually designing the language grounding or curating massive datasets that associate language with the environment. We propose Ground-Compose-Reinforce, an end-to-end, neurosymbolic framework for training RL agents directly from high-level task specifications--without manually designed reward functions or other domain-specific oracles, and without massive datasets. These task specifications take the form of Reward Machines, automata-based representations that capture high-level task structure and are in some cases autoformalizable from natural language. Critically, we show that Reward Machines can be grounded using limited data by exploiting compositionality. Experiments in a custom Meta-World domain with only 350 labelled pretraining trajectories show that our framework faithfully elicits complex behaviours from high-level specifications--including behaviours that never appear in pretraining--while non-compositional approaches fail.

Rodrigo Toro Icarte, Ethan Waldie, Toryn Q. Klassen et al.

Reinforcement learning (RL) is a central problem in artificial intelligence. This problem consists of defining artificial agents that can learn optimal behaviour by interacting with an environment -- where the optimal behaviour is defined with respect to a reward signal that the agent seeks to maximize. Reward machines (RMs) provide a structured, automata-based representation of a reward function that enables an RL agent to decompose an RL problem into structured subproblems that can be efficiently learned via off-policy learning. Here we show that RMs can be learned from experience, instead of being specified by the user, and that the resulting problem decomposition can be used to effectively solve partially observable RL problems. We pose the task of learning RMs as a discrete optimization problem where the objective is to find an RM that decomposes the problem into a set of subproblems such that the combination of their optimal memoryless policies is an optimal policy for the original problem. We show the effectiveness of this approach on three partially observable domains, where it significantly outperforms A3C, PPO, and ACER, and discuss its advantages, limitations, and broader potential.

Parand Alizadeh Alamdari, Toryn Q. Klassen, Rodrigo Toro Icarte et al.

Recent work in AI safety has highlighted that in sequential decision making, objectives are often underspecified or incomplete. This gives discretion to the acting agent to realize the stated objective in ways that may result in undesirable outcomes. We contend that to learn to act safely, a reinforcement learning (RL) agent should include contemplation of the impact of its actions on the wellbeing and agency of others in the environment, including other acting agents and reactive processes. We endow RL agents with the ability to contemplate such impact by augmenting their reward based on expectation of future return by others in the environment, providing different criteria for characterizing impact. We further endow these agents with the ability to differentially factor this impact into their decision making, manifesting behavior that ranges from self-centred to self-less, as demonstrated by experiments in gridworld environments.

Rodrigo Toro Icarte, Toryn Q. Klassen, Richard Valenzano et al.

Reinforcement learning (RL) methods usually treat reward functions as black boxes. As such, these methods must extensively interact with the environment in order to discover rewards and optimal policies. In most RL applications, however, users have to program the reward function and, hence, there is the opportunity to make the reward function visible -- to show the reward function's code to the RL agent so it can exploit the function's internal structure to learn optimal policies in a more sample efficient manner. In this paper, we show how to accomplish this idea in two steps. First, we propose reward machines, a type of finite state machine that supports the specification of reward functions while exposing reward function structure. We then describe different methodologies to exploit this structure to support learning, including automated reward shaping, task decomposition, and counterfactual reasoning with off-policy learning. Experiments on tabular and continuous domains, across different tasks and RL agents, show the benefits of exploiting reward structure with respect to sample efficiency and the quality of resultant policies. Finally, by virtue of being a form of finite state machine, reward machines have the expressive power of a regular language and as such support loops, sequences and conditionals, as well as the expression of temporally extended properties typical of linear temporal logic and non-Markovian reward specification.

Rodrigo Toro Icarte, Richard Valenzano, Toryn Q. Klassen et al.

Reinforcement Learning (RL) agents typically learn memoryless policies---policies that only consider the last observation when selecting actions. Learning memoryless policies is efficient and optimal in fully observable environments. However, some form of memory is necessary when RL agents are faced with partial observability. In this paper, we study a lightweight approach to tackle partial observability in RL. We provide the agent with an external memory and additional actions to control what, if anything, is written to the memory. At every step, the current memory state is part of the agent's observation, and the agent selects a tuple of actions: one action that modifies the environment and another that modifies the memory. When the external memory is sufficiently expressive, optimal memoryless policies yield globally optimal solutions. Unfortunately, previous attempts to use external memory in the form of binary memory have produced poor results in practice. Here, we investigate alternative forms of memory in support of learning effective memoryless policies. Our novel forms of memory outperform binary and LSTM-based memory in well-established partially observable domains.

Maayan Shvo, Toryn Q. Klassen, Sheila A. McIlraith

Theory of Mind is commonly defined as the ability to attribute mental states (e.g., beliefs, goals) to oneself, and to others. A large body of previous work - from the social sciences to artificial intelligence - has observed that Theory of Mind capabilities are central to providing an explanation to another agent or when explaining that agent's behaviour. In this paper, we build and expand upon previous work by providing an account of explanation in terms of the beliefs of agents and the mechanism by which agents revise their beliefs given possible explanations. We further identify a set of desiderata for explanations that utilize Theory of Mind. These desiderata inform our belief-based account of explanation.