Robin Cohen

AAAI Conference 2024 Conference Paper

Multi-Modal Discussion Transformer: Integrating Text, Images and Graph Transformers to Detect Hate Speech on Social Media

• Liam Hebert
• Gaurav Sahu
• Yuxuan Guo
• Nanda Kishore Sreenivas
• Lukasz Golab
• Robin Cohen

We present the Multi-Modal Discussion Transformer (mDT), a novel method for detecting hate speech on online social networks such as Reddit discussions. In contrast to traditional comment-only methods, our approach to labelling a comment as hate speech involves a holistic analysis of text and images grounded in the discussion context. This is done by leveraging graph transformers to capture the contextual relationships in the discussion surrounding a comment and grounding the interwoven fusion layers that combine text and image embeddings instead of processing modalities separately. To evaluate our work, we present a new dataset, HatefulDiscussions, comprising complete multi-modal discussions from multiple online communities on Reddit. We compare the performance of our model to baselines that only process individual comments and conduct extensive ablation studies.

AAMAS Conference 2023 Conference Paper

FedFormer: Contextual Federation with Attention in Reinforcement Learning

• Liam Hebert
• Lukasz Golab
• Pascal Poupart
• Robin Cohen

A core issue in multi-agent federated reinforcement learning is defining how to aggregate insights from multiple agents. This is commonly done by taking the average of each participating agent’s model weights into one common model (FedAvg). We instead propose FedFormer, a novel federation strategy that utilizes Transformer Attention to contextually aggregate embeddings from models originating from different learner agents. In so doing, we attentively weigh the contributions of other agents with respect to the current agent’s environment and learned relationships, thus providing a more effective and efficient federation. We evaluate our methods on the Meta-World environment and find that our approach yields significant improvements over FedAvg and nonfederated Soft Actor-Critic single-agent methods. Our results compared to Soft Actor-Critic show that FedFormer achieves higher episodic return while still abiding by the privacy constraints of federated learning. Finally, we also demonstrate improvements in effectiveness with increased agent pools across all methods in certain tasks. This is contrasted by FedAvg, which fails to make noticeable improvements when scaled.

AAAI Conference 2020 Short Paper

Personalized Prediction of Trust Links in Social Networks (Student Abstract)

• Alexandre Parmentier
• Robin Cohen

In this paper we show how integrating both domain specific and generic trust indicators into a prediction of trust links between users in social networks can improve upon methods for recommending content to users and how clustering of users to deliver personalized solutions offers even greater advantages.

JAAMAS Journal 2019 Journal Article

Inferring true voting outcomes in homophilic social networks

• John A. Doucette
• Alan Tsang
• Robin Cohen

Abstract We investigate the problem of binary opinion aggregation in a social network regarding an objective outcome. Agents receive independent noisy signals relating to the outcome, but may converse with their neighbors in the network before opinions are aggregated, resulting in incorrect opinions gaining prominence in the network. Recent work has shown that, in the general case, there is no procedure for inferring the correct outcome that incorporates information from the connections between agents (i. e. the structure of the social network). We develop a new approach for inferring the true outcome that can benefit from the additional information provided by the social network, under the simple assumption that agents will more readily convert to the true opinion than to a false one, generating a homophilic effect for voters with the correct opinion. Our proposed approach is computationally efficient, and provides significantly more accurate inference in many domains, which we demonstrate via both simulated and real-world datasets. We also theoretically characterize the properties that are necessary for our approach to perform well. Finally, we extend our approach to directed social networks, and cases with many alternatives, and outline areas for future research.

AAMAS Conference 2019 Conference Paper

Trusted AI and the Contribution of Trust Modeling in Multiagent Systems

• Robin Cohen
• Mike Schaekermann
• Sihao Liu
• Michael Cormier

Researchers in the field of artificial intelligence today are increasingly concerned with whether the systems which they build will be “trusted AI", in other words, whether they will be accepted by their human users. The claim of this paper is that these researchers should be aware of the rich set of solutions being developed in the multiagent systems subfield of trust modeling. We propose a specific perspective on how to leverage trust modeling solutions towards assurances for trusted artificial intelligence. We conclude by advocating for greater dialogue between these AI communities.

JAAMAS Journal 2018 Journal Article

Investigating the characteristics of one-sided matching mechanisms under various preferences and risk attitudes

• Hadi Hosseini
• Kate Larson
• Robin Cohen

Abstract One-sided matching mechanisms are fundamental for assigning a set of indivisible objects to a set of self-interested agents when monetary transfers are not allowed. Two widely-studied randomized mechanisms in multiagent settings are the Random Serial Dictatorship (RSD) and the Probabilistic Serial Rule (PS). Both mechanisms require only that agents specify ordinal preferences and have a number of desirable economic and computational properties. However, the induced outcomes of the mechanisms are often incomparable and thus there are challenges when it comes to deciding which mechanism to adopt in practice. In this paper, we first consider the space of general ordinal preferences and provide empirical results on the (in)comparability of RSD and PS. We analyze their respective economic properties under general and lexicographic preferences. We then instantiate utility functions with the goal of gaining insights on the manipulability, efficiency, and envyfreeness of the mechanisms under different risk-attitude models. Our results hold under various preference distribution models, which further confirm the broad use of RSD in most practical applications.

AAMAS Conference 2017 Conference Paper

A Restricted Markov Tree Model for Inference and Generation in Social Choice with Incomplete Preferences

• John A. Doucette
• Robin Cohen

AAMAS Conference 2016 Conference Paper

Investigating the Characteristics of One-Sided Matching Mechanisms (Extended Abstract)

• Hadi Hosseini
• Kate Larson
• Robin Cohen

For one-sided matching problems, two widely studied mechanisms are the Random Serial Dictatorship (RSD) and the Probabilistic Serial Rule (PS). The induced outcomes of these two mechanisms are often incomparable and thus there are challenges when it comes to deciding which mechanism to adopt in practice. Working in the space of general preferences, we provide empirical results on the (in)comparability of RSD and PS and analyze their economic properties.

TIST Journal 2016 Journal Article

Multiagent Resource Allocation for Dynamic Task Arrivals with Preemption

• John A. Doucette
• Graham Pinhey
• Robin Cohen

In this article, we present a distributed algorithm for allocating resources to tasks in multiagent systems, one that adapts well to dynamic task arrivals where new work arises at short notice. Our algorithm is designed to leverage preemption if it is available, revoking resource allocations to tasks in progress if new opportunities arise that those resources are better suited to handle. Our multiagent model assigns a task agent to each task that must be completed and a proxy agent to each resource that is available. Preemption occurs when a task agent approaches a proxy agent with a sufficiently compelling need that the proxy agent determines the newcomer derives more benefit from the proxy agent’s resource than the task agent currently using that resource. Task agents reason about which resources to request based on a learning of churn and congestion. We compare to a well-established multiagent resource allocation framework that permits preemption under more conservative assumptions and show through simulation that our model allows for improved allocations through more permissive preemption. In all, we offer a novel approach for multiagent resource allocation that is able to cope well with dynamic task arrivals.

AAAI Conference 2015 Conference Paper

Conventional Machine Learning for Social Choice

• John Doucette
• Kate Larson
• Robin Cohen

Deciding the outcome of an election when voters have provided only partial orderings over their preferences requires voting rules that accommodate missing data. While existing techniques, including considerable recent work, address missingness through circumvention, we propose the novel application of conventional machine learning techniques to predict the missing components of ballots via latent patterns in the information that voters are able to provide. We show that suitable predictive features can be extracted from the data, and demonstrate the high performance of our new framework on the ballots from many real world elections, including comparisons with existing techniques for voting with partial orderings. Our technique offers a new and interesting conceptualization of the problem, with stronger connections to machine learning than conventional social choice techniques.

TIST Journal 2015 Journal Article

Empowering Patients and Caregivers to Manage Healthcare Via Streamlined Presentation of Web Objects Selected by Modeling Learning Benefits Obtained by Similar Peers

• John Champaign
• Robin Cohen
• Disney Yan Lam

In this article, we introduce a framework for selecting web objects (texts, videos, simulations) from a large online repository to present to patients and caregivers, in order to assist in their healthcare. Motivated by the paradigm of peer-based intelligent tutoring, we model the learning gains achieved by users when exposed to specific web objects in order to recommend those objects most likely to deliver benefit to new users. We are able to show that this streamlined presentation leads to effective knowledge gains, both through a process of simulated learning and through a user study, for the specific application of caring for children with autism. The value of our framework for peer-driven content selection of health information is emphasized through two additional roles for peers: attaching commentary to web objects and proposing subdivided objects for presentation, both of which are demonstrated to deliver effective learning gains, in simulations. In all, we are offering an opportunity for patients to navigate the deep waters of excessive online information towards effective management of healthcare, through content selection influenced by previous peer experiences.

AAAI Conference 2015 Conference Paper

Matching with Dynamic Ordinal Preferences

• Hadi Hosseini
• Kate Larson
• Robin Cohen

We consider the problem of repeatedly matching a set of alternatives to a set of agents with dynamic ordinal preferences. Despite a recent focus on designing one-shot matching mechanisms in the absence of monetary transfers, little study has been done on strategic behavior of agents in sequential assignment problems. We formulate a generic dynamic matching problem via a sequential stochastic matching process. We design a mechanism based on random serial dictatorship (RSD) that, given any history of preferences and matching decisions, guarantees global stochastic strategyproofness while satisfying desirable local properties. We further investigate the notion of envyfreeness in such sequential settings.

AAAI Conference 2015 Conference Paper

On Manipulablity of Random Serial Dictatorship in Sequential Matching with Dynamic Preferences

• Hadi Hosseini
• Kate Larson
• Robin Cohen

We consider the problem of repeatedly matching a set of alternatives to a set of agents in the absence of monetary transfer. We propose a generic framework for evaluating sequential matching mechanisms with dynamic preferences, and show that unlike single-shot settings, the random serial dictatorship mechanism is manipulable.

TIST Journal 2013 Journal Article

A framework for trust modeling in multiagent electronic marketplaces with buying advisors to consider varying seller behavior and the limiting of seller bids

• Jie Zhang
• Robin Cohen

In this article, we present a framework of use in electronic marketplaces that allows buying agents to model the trustworthiness of selling agents in an effective way, making use of seller ratings provided by other buying agents known as advisors. The trustworthiness of the advisors is also modeled, using an approach that combines both personal and public knowledge and allows the relative weighting to be adjusted over time. Through a series of experiments that simulate e-marketplaces, including ones where sellers may vary their behavior over time, we are able to demonstrate that our proposed framework delivers effective seller recommendations to buyers, resulting in important buyer profit. We also propose limiting seller bids as a method for promoting seller honesty, thus facilitating successful selection of sellers by buyers, and demonstrate the value of this approach through experimental results. Overall, this research is focused on the technological aspects of electronic commerce and specifically on technology that would be used to manage trust.

TIST Journal 2013 Journal Article

Validation of an ontological medical decision support system for patient treatment using a repository of patient data

• Atif Khan
• John A. Doucette
• Robin Cohen

In this article, we begin by presenting OMeD, a medical decision support system, and argue for its value over purely probabilistic approaches that reason about patients for time-critical decision scenarios. We then progress to present Holmes, a Hybrid Ontological and Learning MEdical System which supports decision making about patient treatment. This system is introduced in order to cope with the case of missing data. We demonstrate its effectiveness by operating on an extensive set of real-world patient health data from the CDC, applied to the decision-making scenario of administering sleeping pills. In particular, we clarify how the combination of semantic, ontological representations, and probabilistic reasoning together enable the proposal of effective patient treatments. Our focus is thus on presenting an approach for interpreting medical data in the context of real-time decision making. This constitutes a comprehensive framework for the design of medical recommendation systems for potential use by medical professionals and patients both, with the end result being personalized patient treatment. We conclude with a discussion of the value of our particular approach for such diverse considerations as coping with misinformation provided by patients, performing effectively in time-critical environments where real-time decisions are necessary, and potential applications facilitating patient information gathering.

AAAI Conference 2012 Conference Paper

A Market-Based Coordination Mechanism for Resource Planning Under Uncertainty

• Hadi Hosseini
• Jesse Hoey
• Robin Cohen

AAMAS Conference 2012 Conference Paper

Detecting and Identifying Coalitions

• Reid Kerr
• Robin Cohen

In multiagent scenarios, subsets of a population (coalitions) may attempt to cooperate, for mutual benefit. We present a technique for detecting the presence of coalitions (malicious or otherwise) and identifying their members, and demonstrate its effectiveness.

IJCAI Conference 2009 Conference Paper

• Georgia Kastidou
• Kate Larson
• Robin Cohen

We introduce a framework so that communities can exchange reputation information about agents in environments where agents are migrating between communities. We view the acquisition of the reputation information as a purchase and focus on the design of a payment function to facilitate the payment for information in a way that motivates communities to truthfully report reputation information for agents. We prove that in our proposed framework, honesty is the optimal policy and demonstrate the value of using a payment-function approach for the exchange of reputation information about agents between communities in multiagent environments. Using our payment function, each community is strengthened: it is able to reason more effectively about which agents to accept and can enjoy agents that are motivated to contribute strongly to the benefit of the community.

AAMAS Conference 2009 Conference Paper

Smart Cheaters Do Prosper: Defeating Trust and Reputation Systems

• Reid Kerr
• Robin Cohen

Traders in electronic marketplaces may behave dishonestly, cheating other agents. A multitude of trust and reputation systems have been proposed to try to cope with the problem of cheating. These systems are often evaluated by measuring their performance against simple agents that cheat randomly. Unfortunately, these systems are not often evaluated from the perspective of security—can a motivated attacker defeat the protection? Previously, it was argued that existing systems may suffer from vulnerabilities that permit effective, profitable cheating despite the use of the system. In this work, we experimentally substantiate the presence of these vulnerabilities by successfully implementing and testing a number of such ‘attacks’, which consist only of sequences of sales (honest and dishonest) that can be executed in the system. This investigation also reveals two new, previously-unnoted cheating techniques. Our success in executing these attacks compellingly makes a key point: security must be a central design goal for developers of trust and reputation systems.

AAMAS Conference 2007 Conference Paper

An Incentive Mechanism for Eliciting Fair Ratings of Sellers in E-Marketplaces

• Jie Zhang
• Robin Cohen

In this paper, we propose a novel incentive mechanism for eliciting fair ratings of selling agents from buying agents. In our mechanism, buyers model other buyers and select the most trustworthy ones as their neighbors from whom they can ask advice about sellers. In addition, however, sellers model the reputation of buyers. Reputable buyers always provide fair ratings of sellers, and are likely to be neighbors of many other buyers. In marketplaces operating with our mechanism, sellers will increase quality and decrease prices of products to satisfy reputable buyers. In consequence, our mechanism creates incentives for buyers to provide fair ratings of sellers.

AAMAS Conference 2007 Conference Paper

Towards Provably Secure Trust and Reputation Systems in E-Marketplaces

• Reid Kerr
• Robin Cohen

In this paper, we present a framework for evaluating the security of trust and reputation systems for electronic marketplaces populated with buying and selling agents. Our proposed framework offers a method for researchers to understand the security of their systems, and to provide precise guarantees of the degree of provable security that these systems offer. We demonstrate the viability of our proposed framework by presenting a specific monetarybased trust system known as Trunits, along with an analysis that shows that Trunits provides a guaranteed level of security for buyers.

AAMAS Conference 2004 Conference Paper

Improving User Satisfaction in Agent-Based Electronic Marketplaces by Reputation Modelling and Adjus

• Thomas Tran
• Robin Cohen

IJCAI Conference 2003 Conference Paper

Learning Algorithms for Software Agents in Uncertain and Untrusted Market Environments

• Thomas Tran
• Robin Cohen

IJCAI Conference 1995 Conference Paper

mproving Heuristic Based Temporal Analysis of Narratives wLth Aspect Determination

• Fel Song
• Robin Cohen

IJCAI Conference 1991 Conference Paper

Resolving Plan Ambiguity for Cooperative Response Generation

• Peter van Beek
• Robin Cohen

AAAI Conference 1987 Conference Paper

Interpreting Clues in Conjunction with Processing Restrictions in Arguments and Discourse

• Robin Cohen

This paper extends previous work which provided a theory for the interpretation of and necessity for clue words in a particular kind of discourse - namely, one-way arguments. Previous work described a taxonomy of connective clues (words such as "hence" or phrases such as "as a result"), where each clue, classified according to the taxonomy, would set in place a default interpretation of its containing proposition, with respect to the representation for the argument so far. In this paper, we examine how to combine the restrictions for clues with a basic processor for the discourse, offering a integrated processing algorithm, which takes advantage of clues to reduce processing and to detect incoherent arguments, and can still produce an analysis in the absence of clues. We conclude with some suggestions for incorporating clues of re-direction and clues that signal exceptional transmissions. We also demonstrate the implications of our results for discourse in general.