Gourab K Patro

Gourab K. Patro, Prithwish Jana, Abhijnan Chakraborty et al. · gatech

Recently, almost all conferences have moved to virtual mode due to the pandemic-induced restrictions on travel and social gathering. Contrary to in-person conferences, virtual conferences face the challenge of efficiently scheduling talks, accounting for the availability of participants from different timezones and their interests in attending different talks. A natural objective for conference organizers is to maximize efficiency, e.g., total expected audience participation across all talks. However, we show that optimizing for efficiency alone can result in an unfair virtual conference schedule, where individual utilities for participants and speakers can be highly unequal. To address this, we formally define fairness notions for participants and speakers, and derive suitable objectives to account for them. As the efficiency and fairness objectives can be in conflict with each other, we propose a joint optimization framework that allows conference organizers to design schedules that balance (i.e., allow trade-offs) among efficiency, participant fairness and speaker fairness objectives. While the optimization problem can be solved using integer programming to schedule smaller conferences, we provide two scalable techniques to cater to bigger conferences. Extensive evaluations over multiple real-world datasets show the efficacy and flexibility of our proposed approaches.

Niloy Ganguly, Dren Fazlija, Maryam Badar et al.

State-of-the-art AI models largely lack an understanding of the cause-effect relationship that governs human understanding of the real world. Consequently, these models do not generalize to unseen data, often produce unfair results, and are difficult to interpret. This has led to efforts to improve the trustworthiness aspects of AI models. Recently, causal modeling and inference methods have emerged as powerful tools. This review aims to provide the reader with an overview of causal methods that have been developed to improve the trustworthiness of AI models. We hope that our contribution will motivate future research on causality-based solutions for trustworthy AI.

Himanshu Gharat, Himanshi Agrawal, Gourab K. Patro

Large Language Models (LLMs) have empowered AI agents with advanced capabilities for understanding, reasoning, and interacting across diverse tasks. The addition of memory further enhances them by enabling continuity across interactions, learning from past experiences, and improving the relevance of actions and responses over time; termed as memory-enhanced personalization. Although such personalization through memory offers clear benefits, it also introduces risks of bias. While several previous studies have highlighted bias in ML and LLMs, bias due to memory-enhanced personalized agents is largely unexplored. Using recruitment as an example use case, we simulate the behavior of a memory-enhanced personalized agent, and study whether and how bias is introduced and amplified in and across various stages of operation. Our experiments on agents using safety-trained LLMs reveal that bias is systematically introduced and reinforced through personalization, emphasizing the need for additional protective measures or agent guardrails in memory-enhanced LLM-based AI agents.

Hailay Teklehaymanot, Dren Fazlija, Niloy Ganguly et al.

The absence of explicitly tailored, accessible annotated datasets for educational purposes presents a notable obstacle for NLP tasks in languages with limited resources.This study initially explores the feasibility of using machine translation (MT) to convert an existing dataset into a Tigrinya dataset in SQuAD format. As a result, we present TIGQA, an expert annotated educational dataset consisting of 2.68K question-answer pairs covering 122 diverse topics such as climate, water, and traffic. These pairs are from 537 context paragraphs in publicly accessible Tigrinya and Biology books. Through comprehensive analyses, we demonstrate that the TIGQA dataset requires skills beyond simple word matching, requiring both single-sentence and multiple-sentence inference abilities. We conduct experiments using state-of-the art MRC methods, marking the first exploration of such models on TIGQA. Additionally, we estimate human performance on the dataset and juxtapose it with the results obtained from pretrained models.The notable disparities between human performance and best model performance underscore the potential for further enhancements to TIGQA through continued research. Our dataset is freely accessible via the provided link to encourage the research community to address the challenges in the Tigrinya MRC.

Gourab K Patro, Himanshi Agrawal, Himanshu Gharat et al.

Modern general-purpose AI systems made using large language and vision models, are capable of performing a range of tasks like writing text articles, generating and debugging codes, querying databases, and translating from one language to another, which has made them quite popular across industries. However, there are risks like hallucinations, toxicity, and stereotypes in their output that make them untrustworthy. We review various risks and vulnerabilities of modern general-purpose AI along eight widely accepted responsible AI (RAI) principles (fairness, privacy, explainability, robustness, safety, truthfulness, governance, and sustainability) and compare how they are non-existent or less severe and easily mitigable in traditional task-specific counterparts. We argue that this is due to the non-deterministically high Degree of Freedom in output (DoFo) of general-purpose AI (unlike the deterministically constant or low DoFo of traditional task-specific AI systems), and there is a need to rethink our approach to RAI for general-purpose AI. Following this, we derive C2V2 (Control, Consistency, Value, Veracity) desiderata to meet the RAI requirements for future general-purpose AI systems, and discuss how recent efforts in AI alignment, retrieval-augmented generation, reasoning enhancements, etc. fare along one or more of the desiderata. We believe that the goal of developing responsible general-purpose AI can be achieved by formally modeling application- or domain-dependent RAI requirements along C2V2 dimensions, and taking a system design approach to suitably combine various techniques to meet the desiderata.

Gourab K Patro, Lorenzo Porcaro, Laura Mitchell et al.

Ranking, recommendation, and retrieval systems are widely used in online platforms and other societal systems, including e-commerce, media-streaming, admissions, gig platforms, and hiring. In the recent past, a large "fair ranking" research literature has been developed around making these systems fair to the individuals, providers, or content that are being ranked. Most of this literature defines fairness for a single instance of retrieval, or as a simple additive notion for multiple instances of retrievals over time. This work provides a critical overview of this literature, detailing the often context-specific concerns that such an approach misses: the gap between high ranking placements and true provider utility, spillovers and compounding effects over time, induced strategic incentives, and the effect of statistical uncertainty. We then provide a path forward for a more holistic and impact-oriented fair ranking research agenda, including methodological lessons from other fields and the role of the broader stakeholder community in overcoming data bottlenecks and designing effective regulatory environments.

Arpita Biswas, Gourab K Patro, Niloy Ganguly et al.

Many online platforms today (such as Amazon, Netflix, Spotify, LinkedIn, and AirBnB) can be thought of as two-sided markets with producers and customers of goods and services. Traditionally, recommendation services in these platforms have focused on maximizing customer satisfaction by tailoring the results according to the personalized preferences of individual customers. However, our investigation reinforces the fact that such customer-centric design of these services may lead to unfair distribution of exposure to the producers, which may adversely impact their well-being. On the other hand, a pure producer-centric design might become unfair to the customers. As more and more people are depending on such platforms to earn a living, it is important to ensure fairness to both producers and customers. In this work, by mapping a fair personalized recommendation problem to a constrained version of the problem of fairly allocating indivisible goods, we propose to provide fairness guarantees for both sides. Formally, our proposed {\em FairRec} algorithm guarantees Maxi-Min Share ($α$-MMS) of exposure for the producers, and Envy-Free up to One Item (EF1) fairness for the customers. Extensive evaluations over multiple real-world datasets show the effectiveness of {\em FairRec} in ensuring two-sided fairness while incurring a marginal loss in overall recommendation quality. Finally, we present a modification of FairRec (named as FairRecPlus) that at the cost of additional computation time, improves the recommendation performance for the customers, while maintaining the same fairness guarantees.

Ashmi Banerjee, Gourab K Patro, Linus W. Dietz et al.

The proliferation of smartphones has led to the increased popularity of location-based search and recommendation systems. Online platforms like Google and Yelp allow location-based search in the form of nearby feature to query for hotels or restaurants in the vicinity. Moreover, hotel booking platforms like Booking[dot]com, Expedia, or Trivago allow travelers searching for accommodations using either their desired location as a search query or near a particular landmark. Since the popularity of different locations in a city varies, certain locations may get more queries than other locations. Thus, the exposure received by different establishments at these locations may be very different from their intrinsic quality as captured in their ratings. Today, many small businesses (shops, hotels, or restaurants) rely on such online platforms for attracting customers. Thus, receiving less exposure than that is expected can be unfavorable for businesses. It could have a negative impact on their revenue and potentially lead to economic starvation or even shutdown. By gathering and analyzing data from three popular platforms, we observe that many top-rated hotels and restaurants get less exposure vis-a-vis their quality, which could be detrimental for them. Following a meritocratic notion, we define and quantify such exposure disparity due to location-based searches on these platforms. We attribute this exposure disparity mainly to two kinds of biases -- Popularity Bias and Position Bias. Our experimental evaluation on multiple datasets reveals that although the platforms are doing well in delivering distance-based results, exposure disparity exists for individual businesses and needs to be reduced for business sustainability.

Aadi Swadipto Mondal, Rakesh Bal, Sayan Sinha et al.

Recommender systems are one of the most widely used services on several online platforms to suggest potential items to the end-users. These services often use different machine learning techniques for which fairness is a concerning factor, especially when the downstream services have the ability to cause social ramifications. Thus, focusing on the non-personalised (global) recommendations in news media platforms (e.g., top-k trending topics on Twitter, top-k news on a news platform, etc.), we discuss on two specific fairness concerns together (traditionally studied separately)---user fairness and organisational fairness. While user fairness captures the idea of representing the choices of all the individual users in the case of global recommendations, organisational fairness tries to ensure politically/ideologically balanced recommendation sets. This makes user fairness a user-side requirement and organisational fairness a platform-side requirement. For user fairness, we test with methods from social choice theory, i.e., various voting rules known to better represent user choices in their results. Even in our application of voting rules to the recommendation setup, we observe high user satisfaction scores. Now for organisational fairness, we propose a bias metric which measures the aggregate ideological bias of a recommended set of items (articles). Analysing the results obtained from voting rule-based recommendation, we find that while the well-known voting rules are better from the user side, they show high bias values and clearly not suitable for organisational requirements of the platforms. Thus, there is a need to build an encompassing mechanism by cohesively bridging ideas of user fairness and organisational fairness. In this abstract paper, we intend to frame the elementary ideas along with the clear motivation behind the requirement of such a mechanism.

Gourab K Patro, Abhijnan Chakraborty, Niloy Ganguly et al.

The (COVID-19) pandemic-induced restrictions on travel and social gatherings have prompted most conference organizers to move their events online. However, in contrast to physical conferences, virtual conferences face a challenge in efficiently scheduling talks, accounting for the availability of participants from different time-zones as well as their interests in attending different talks. In such settings, a natural objective for the conference organizers would be to maximize some global welfare measure, such as the total expected audience participation across all talks. However, we show that optimizing for global welfare could result in a schedule that is unfair to the stakeholders, i.e., the individual utilities for participants and speakers can be highly unequal. To address the fairness concerns, we formally define fairness notions for participants and speakers, and subsequently derive suitable fairness objectives for them. We show that the welfare and fairness objectives can be in conflict with each other, and there is a need to maintain a balance between these objective while caring for them simultaneously. Thus, we propose a joint optimization framework that allows conference organizers to design talk schedules that balance (i.e., allow trade-offs) between global welfare, participant fairness and the speaker fairness objectives. We show that the optimization problem can be solved using integer linear programming, and empirically evaluate the necessity and benefits of such joint optimization approach in virtual conference scheduling.

Jessie Finocchiaro, Roland Maio, Faidra Monachou et al.

Decision-making systems increasingly orchestrate our world: how to intervene on the algorithmic components to build fair and equitable systems is therefore a question of utmost importance; one that is substantially complicated by the context-dependent nature of fairness and discrimination. Modern decision-making systems that involve allocating resources or information to people (e.g., school choice, advertising) incorporate machine-learned predictions in their pipelines, raising concerns about potential strategic behavior or constrained allocation, concerns usually tackled in the context of mechanism design. Although both machine learning and mechanism design have developed frameworks for addressing issues of fairness and equity, in some complex decision-making systems, neither framework is individually sufficient. In this paper, we develop the position that building fair decision-making systems requires overcoming these limitations which, we argue, are inherent to each field. Our ultimate objective is to build an encompassing framework that cohesively bridges the individual frameworks of mechanism design and machine learning. We begin to lay the ground work towards this goal by comparing the perspective each discipline takes on fair decision-making, teasing out the lessons each field has taught and can teach the other, and highlighting application domains that require a strong collaboration between these disciplines.