United Kingdom

In this work, we propose a framework that enables a human to teach a robot a new task by interactively providing it with unlabeled instructions. We ground the meaning of instruction signals in the task-learning process, and use them simultaneously for guiding the latter. We implement our framework as a modular architecture, named TICS (Task-Instruction-Contingency-Shaping) that combines different information sources: a predefined reward function, human evaluative feedback and unlabeled instructions. This approach provides a novel perspective for robotic task learning that lies between Reinforcement Learning and Supervised Learning paradigms. We evaluate our framework both in simulation and with a real robot. The experimental results demonstrate the effectiveness of our framework in accelerating the task-learning process and in reducing the number of required teaching signals.

AAMAS 2021

Teaching a Robot with Unlabeled Instructions: The TICS Architecture (JAAMAS Track)

AAMAS is the largest and most influential conference in the area of agents and multiagent systems, bringing together researchers and practitioners in all areas of agent technology and providing and internationally renowned high-profile forum for publishing and finding out about the latest developments in the field. AAMAS is the flagship conference of the non-profit International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS).  

![](https://assets.underline.io/uploads/markdown_image/1/image/ce4604bd54468a006164d16e080204fb.jpg)

Please click on the button below to register for this event.

If you are registered and see this page please check the account email you are logged in with. It must be the same as for the registration.

If you still have trouble accessing the contact please contact support at aamas2021@underline.io 

technical paper

Given their installation on nearly a billion consumer devices around the world, consumers clearly enjoy using voice-driven assistants such as Alexa, Siri and Google Home, and find it compelling to interact with AI agents as quasi-human entities. Inevitably, people who are accustomed to using voice-driven assistants at home and in the car will expect to use such technologies in the workplace. What form will this take? Simple extrapolations from consumer space (e.g. running meetings or presentations) promise only modest value. I propose that AAMAS and the AI research community should pursue a bolder vision in which software agents act as quasi-human collaborators on core business intelligence tasks that entail analyzing data, diagnosing problems, and making decisions. Moreover, these business intelligence agents should communicate multi-modally, i.e. they must understand and employ speech complemented by non-verbal behaviors such as pointing, eye gaze, and facial expression. I outline requisite agent, MAS and AI technologies and pose several fundamental research challenges raised by this vision.

Multi-modal Agents for Business Intelligence (Blue Sky Ideas Track)

To develop and effectively deploy Trustworthy Autonomous Systems (TAS), we face various social, technological, legal, and ethical challenges in which different notions of responsibility can play a key role. In this work, we elaborate on these challenges, discuss research gaps, and show how the multidimensional notion of responsibility can play a role to bridge them. We argue that TAS requires operational tools to represent and reason about responsibilities of humans as well as AI agents. We review major challenges to which responsibility reasoning can contribute, highlight open research problems, and argue for the application of multiagent responsibility models in a variety of TAS domains.

Responsibility Research for Trustworthy Autonomous Systems (Blue Sky Ideas Track)

This paper discusses the formalization of a norm enforcement mechanism for a regulative time constrained conditional normative framework. The norm representation captures both the condition and effect of a norm as situations and allows for rich temporal constraints between the times of the situations. As part of the enforcement process, a designated agent has an obligation within a time constraint to inform a liable agent of a reparative action they must take and the time constraint within which they must take it. That same agent must then monitor the compliance of the erring agent’s obligation to carry out the reparative action, meting out sanctions in the case of violation.

A Norm Enforcement Mechanism for a Time-Constrained Conditional Normative Framework (JAAMAS Track)

It is claimed that, in the nascent 'Cognitive Era', intelligent systems will be trained using machine learning techniques rather than programmed by software developers. A contrary point of view argues that machine learning has limitations, and, taken in isolation, cannot form the basis of autonomous systems capable of intelligent behaviour in complex environments. In this paper, we argue that the unique strengths of Belief-Desire-Intention (BDI) agent programming languages provide an ideal framework for integrating the wide range of AI capabilities necessary for progress towards the next-generation of intelligent systems.

Agent Programming in the Cognitive Era (JAAMAS Track)

A relevant application domain for agent-based software is given by smart mobile applications. In this context, the impressive progress of technologies makes it possible to explore the use of agent-oriented programming languages and frameworks based on cognitive architectures. Accordingly, here we provide a description of JaCa-Android, a framework based on the JaCaMo platform that allows for designing and programming smart mobile apps using BDI-based cognitive agents within the A&A conceptual model.

Programming Agent-based Mobile Apps: The JaCa-Android Framework (JAAMAS Track)

We describe an approach for constraining robot autonomy based on the robot's awareness of patterns of its human teammates' behaviors, rather than either ignoring its teammates (which is fast but dangerous) or inferring their plans (which is safer but slow). We evaluate this approach in a series of simulated problems where an unmanned ground vehicle and its human teammates must rapidly respond to a sudden context shift, and identify conditions that should be (purposely) met such that a pattern-aware approach is particularly effective compared to the alternatives.

On Teammate-Pattern-Aware Autonomy (JAAMAS Track)

Multiagent system models do not typically encompass tools for tackling abnormal or exceptional situations. When a critical situation arises, and individual agents fail, the system as a whole usually fails, too. For realizing robust distributed systems, conceived as agent systems or organizations, it is necessary to keep a right level of situational awareness (of both agents and environment), through the introduction of the means for gathering and propagating feedback, upon which actions can be taken. This work proposes a notion of accountability that encompasses both a normative dimension, and a structural dimension, that serves for the purpose of robustness.

Robustness Based on Accountability in Multiagent Organizations

Community Health Workers (CHWs) form an important component of health-care systems globally, especially in low-resource settings. CHWs are often tasked with monitoring the health of and intervening on their patient cohort. Previous work has developed several classes of Restless Multi-Armed Bandits (RMABs) that are computationally tractable and indexable, a condition that guarantees asymptotic optimality, for solving such health monitoring and intervention problems (HMIPs). However, existing solutions to HMIPs fail to account for risk-sensitivity considerations of CHWs in the planning stage and may run the danger of ignoring some patients completely because they are deemed less valuable to intervene on. Additionally, these also rely on patients reporting their state of adherence accurately when intervened upon. Towards tackling these issues, our contributions in this paper are as follows: (1) We develop an RMAB solution to HMIPs that allows for reward functions that are monotone increasing, rather than linear, in the belief state and also supports a wider class of observations. (2) We prove theoretical guarantees on the asymptotic optimality of our algorithm for any arbitrary reward function. Additionally, we show that for the specific reward function considered in previous work, our theoretical conditions are stronger than the state-of-the-art guarantees. (3) We show the applicability of these new results for addressing the three issues pertaining to: risk-sensitive planning, equitable allocation and reliance on perfect observations as highlighted above. We evaluate these techniques on both simulated as well as real data from a prevalent CHW task of monitoring adherence of tuberculosis patients to their prescribed medication in Mumbai, India and show improved performance over the state-of-the-art. The simulation code is available at: https://github.com/AdityaMate/risk-aware-bandits.

Risk-Aware Interventions in Public Health: Planning with Restless Multi-Armed Bandits

Reliable and secure communication between heterogeneously resourced autonomous agents controlled by competing stakeholders in a decentralized environment is a challenge. Agents require a means to find each other and communicate without reliance on a centralized party and participation in the system must be permissionless. We present the Agent Communication Network (ACN), a peer-to-peer lookup system that provides a distributed overlay to the Internet and addresses this problem. The ACN enables agents to find each other and to communicate safely. It achieves this by leveraging a distributed hash table for agent lookup, maintained by participating peers and through the use of public-key cryptography. The paper discusses the properties of the system and its guarantees as well as its integration with a novel multi-agent system. Preliminary benchmark results demonstrate the feasibility of the system, its performance, and its scalability.

Peer-to-peer Autonomous Agent Communication Network

In multi-agent systems the main process responsible for obtaining information about the environment is perception, generally this process is performed passively regardless the agent's intentional state. However, especially when inserted in the real world, a frequent problem is that agents have partial perception of the environment, failing to perceive some relevant information. To circumvent this problem, a solution is to actively take actions to perceive what is of interest to the agent, for example, in a computer vision system, the camera can be repositioned to have a better view of an object. This work aims to develop an active perception model integrated with the reasoning cycle of BDI agents. Experiments are performed using BDI agents with ROS to command unmanned aerial vehicles to analyze the benefits and impacts of using cognitive agents with active perception to program robot intelligence.

Teaching a Robot with Unlabeled Instructions: The TICS Architecture (JAAMAS Track)

Downloads

Next from AAMAS 2021

Multi-modal Agents for Business Intelligence (Blue Sky Ideas Track)

Similar lecture

Cooperative Prioritized Sweeping

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Teaching a Robot with Unlabeled Instructions: The TICS Architecture (JAAMAS Track)

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AAMAS 2021

Multi-modal Agents for Business Intelligence (Blue Sky Ideas Track)

Similar lecture

Cooperative Prioritized Sweeping

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads