Janusz Marecki

Moucheng Xu, Evangelos Chatzaroulas, Luc McCutcheon et al.

A Standard Operating Procedure (SOP) defines a low-level, step-by-step written guide for a business software workflow. SOP generation is a crucial step towards automating end-to-end software workflows. Manually creating SOPs can be time-consuming. Recent advancements in large video-language models offer the potential for automating SOP generation by analyzing recordings of human demonstrations. However, current large video-language models face challenges with zero-shot SOP generation. In this work, we first explore in-context learning with video-language models for SOP generation. We then propose an exploration-focused strategy called In-Context Ensemble Learning, to aggregate pseudo labels of multiple possible paths of SOPs. The proposed in-context ensemble learning as well enables the models to learn beyond its context window limit with an implicit consistency regularisation. We report that in-context learning helps video-language models to generate more temporally accurate SOP, and the proposed in-context ensemble learning can consistently enhance the capabilities of the video-language models in SOP generation.

Nicolai Dorka, Janusz Marecki, Ammar Anwar

Addressing the challenge of a digital assistant capable of executing a wide array of user tasks, our research focuses on the realm of instruction-based mobile device control. We leverage recent advancements in large language models (LLMs) and present a visual language model (VLM) that can fulfill diverse tasks on mobile devices. Our model functions by interacting solely with the user interface (UI). It uses the visual input from the device screen and mimics human-like interactions, encompassing gestures such as tapping and swiping. This generality in the input and output space allows our agent to interact with any application on the device. Unlike previous methods, our model operates not only on a single screen image but on vision-language sentences created from sequences of past screenshots along with corresponding actions. Evaluating our method on the challenging Android in the Wild benchmark demonstrates its promising efficacy and potential.

Kavya Kopparapu, Edgar A. Duéñez-Guzmán, Jayd Matyas et al.

A key challenge in the study of multiagent cooperation is the need for individual agents not only to cooperate effectively, but to decide with whom to cooperate. This is particularly critical in situations when other agents have hidden, possibly misaligned motivations and goals. Social deduction games offer an avenue to study how individuals might learn to synthesize potentially unreliable information about others, and elucidate their true motivations. In this work, we present Hidden Agenda, a two-team social deduction game that provides a 2D environment for studying learning agents in scenarios of unknown team alignment. The environment admits a rich set of strategies for both teams. Reinforcement learning agents trained in Hidden Agenda show that agents can learn a variety of behaviors, including partnering and voting without need for communication in natural language.

Joel Z. Leibo, Vinicius Zambaldi, Marc Lanctot et al.

Matrix games like Prisoner's Dilemma have guided research on social dilemmas for decades. However, they necessarily treat the choice to cooperate or defect as an atomic action. In real-world social dilemmas these choices are temporally extended. Cooperativeness is a property that applies to policies, not elementary actions. We introduce sequential social dilemmas that share the mixed incentive structure of matrix game social dilemmas but also require agents to learn policies that implement their strategic intentions. We analyze the dynamics of policies learned by multiple self-interested independent learning agents, each using its own deep Q-network, on two Markov games we introduce here: 1. a fruit Gathering game and 2. a Wolfpack hunting game. We characterize how learned behavior in each domain changes as a function of environmental factors including resource abundance. Our experiments show how conflict can emerge from competition over shared resources and shed light on how the sequential nature of real world social dilemmas affects cooperation.

Marek Petrik, Dharmashankar Subramanian, Janusz Marecki

We propose solution methods for previously-unsolved constrained MDPs in which actions can continuously modify the transition probabilities within some acceptable sets. While many methods have been proposed to solve regular MDPs with large state sets, there are few practical approaches for solving constrained MDPs with large action sets. In particular, we show that the continuous action sets can be replaced by their extreme points when the rewards are linear in the modulation. We also develop a tractable optimization formulation for concave reward functions and, surprisingly, also extend it to non- concave reward functions by using their concave envelopes. We evaluate the effectiveness of the approach on the problem of managing delinquencies in a portfolio of loans.

Alan S. Carlin, Nathan Schurr, Janusz Marecki

The Next Generation Air Transportation System will introduce new, advanced sensor technologies into the cockpit. With the introduction of such systems, the responsibilities of the pilot are expected to dramatically increase. In the ALARMS (ALerting And Reasoning Management System) project for NASA, we focus on a key challenge of this environment, the quick and efficient handling of aircraft sensor alerts. It is infeasible to alert the pilot on the state of all subsystems at all times. Furthermore, there is uncertainty as to the true hazard state despite the evidence of the alerts, and there is uncertainty as to the effect and duration of actions taken to address these alerts. This paper reports on the first steps in the construction of an application designed to handle Next Generation alerts. In ALARMS, we have identified 60 different aircraft subsystems and 20 different underlying hazards. In this paper, we show how a Bayesian network can be used to derive the state of the underlying hazards, based on the sensor input. Then, we propose a framework whereby an automated system can plan to address these hazards in cooperation with the pilot, using a Time-Dependent Markov Process (TMDP). Different hazards and pilot states will call for different alerting automation plans. We demonstrate this emerging application of Bayesian networks and TMDPs to cockpit automation, for a use case where a small number of hazards are present, and analyze the resulting alerting automation policies.