Algorithmic systems are often called upon to assist in high-stakes decision making. In light of this, algorithmic recourse, the principle wherein individuals should be able to take action against an undesirable outcome made by an algorithmic system, is receiving growing attention. The bulk of the literature on algorithmic recourse to-date focuses primarily on how to provide recourse to a single individual, overlooking a critical element: the effects of a continuously changing context. Disregarding these effects on recourse is a significant oversight, since, in almost all cases, recourse consists of an individual making a first, unfavorable attempt, and then being given an opportunity to make one or several attempts at a later date — when the context might have changed. This can create false expectations, as initial recourse recommendations may become less reliable over time due to model drift and competition for access to the favorable outcome between individuals.
In this work we propose an agent-based simulation framework for studying the effects of a continuously changing environment on algorithmic recourse. In particular, we identify two main effects that can alter the reliability of recourse for individuals represented by the agents: (1) competition with other agents acting upon recourse, and (2) competition with new agents entering the environment. Our findings highlight that only a small set of specific parameterizations result in algorithmic recourse that is reliable for agents over time. Consequently, we argue that substantial additional work is needed to understand recourse reliability over time, and to develop recourse methods that reward agents’ effort.
more »
« less
Quantifying harm
In earlier work, we defined a qualitative notion of harm: either harm is caused, or it is not.
For practical applications, we often need to quantify harm; for example, we may want to choose the least harmful of a set of possible interventions. We first present a quantitative definition of harm in a deterministic context involving a single individual, then we consider the issues involved in dealing with uncertainty regarding the context and going from a notion of harm for a single individual to a notion of ``societal harm'', which involves aggregating the harm to individuals. We show that the ``obvious'' way
of doing this (just taking the expected harm for an individual and then summing the expected harm over all individuals) can lead to counterintuitive or inappropriate answers, and discuss alternatives, drawing on work from the decision-theory literature.
more »
« less
- Award ID(s):
- 1703846
- PAR ID:
- 10414210
- Date Published:
- Journal Name:
- Proceedings of the 32nd International Joint Conference on Artificial Intelligence (IJCAI 2023)
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Abstract Sequencing high molecular weight (HMW) DNA with long-read and linked-read technologies has promoted a major increase in more complete genome sequences for nonmodel organisms. Sequencing approaches that rely on HMW DNA have been limited to larger organisms or pools of multiple individuals, but recent advances have allowed for sequencing from individuals of small-bodied organisms. Here, we use HMW DNA sequencing with PacBio long reads and TELL-Seq linked reads to assemble and annotate the genome from a single individual feather louse (Brueelia nebulosa) from a European Starling (Sturnus vulgaris). We assembled a genome with a relatively high scaffold N50 (637 kb) and with BUSCO scores (96.1%) comparable to louse genomes assembled from pooled individuals. We annotated a number of genes (10,938) similar to the human louse (Pediculus humanus) genome. Additionally, calling phased variants revealed that the Brueelia genome is more heterozygous (∼1%) then expected for a highly obligate and dispersal-limited parasite. We also assembled and annotated the mitochondrial genome and primary endosymbiont (Sodalis) genome from the individual louse, which showed evidence for heteroplasmy in the mitogenome and a reduced genome size in the endosymbiont compared to its free-living relative. Our study is a valuable demonstration of the capability to obtain high-quality genomes from individual small, nonmodel organisms. Applying this approach to other organisms could greatly increase our understanding of the diversity and evolution of individual genomes.
-
null (Ed.)Diffusion of information in social network has been the focus of intense research in the recent past decades due to its significant impact in shaping public discourse through group/individual influence. Existing research primarily models influence as a binary property of entities: influenced or not influenced. While this is a useful abstraction, it discards the notion of degree of influence, i.e., certain individuals may be influenced ``more'' than others. We introduce the notion of \emph{attitude}, which, as described in social psychology, is the degree by which an entity is influenced by the information. Intuitively, attitude captures the number of distinct neighbors of an entity influencing the latter. We present an information diffusion model (AIC model) that quantifies the degree of influence, i.e., attitude of individuals, in a social network. With this model, we formulate and study attitude maximization problem. We prove that the function for computing attitude is monotonic and sub-modular, and the attitude maximization problem is NP-Hard. We present a greedy algorithm for maximization with an approximation guarantee of $(1-1/e)$. In the context of AIC model, we study two problems, with the aim to investigate the scenarios where attaining individuals with high attitude is objectively more important than maximizing the attitude of the entire network. In the first problem, we introduce the notion of \emph{actionable attitude}; intuitively, individuals with actionable attitude are likely to ``act'' on their attained attitude. We show that the function for computing actionable attitude, unlike that for computing attitude, is non-submodular and however is \emph{approximately submodular}. We present approximation algorithm for maximizing actionable attitude in a network. In the second problem, we consider identifying the number of individuals in the network with attitude above a certain value, a threshold. In this context, the function for computing the number of individuals with attitude above a given threshold induced by a seed set is \emph{neither submodular nor supermodular}. We present heuristics for realizing the solution to the problem. We experimentally evaluated our algorithms and studied empirical properties of the attitude of nodes in network such as spatial and value distribution of high attitude nodes.more » « less
-
Many selection procedures involve ordering candidates according to their qualifications. For example, a university might order applicants according to a perceived probability of graduation within four years, and then select the top 1000 applicants. In this work, we address the problem of ranking members of a population according to their “probability” of success, based on a training set of historical binary outcome data (e.g., graduated in four years or not). We show how to obtain rankings that satisfy a number of desirable accuracy and fairness criteria, despite the coarseness of the training data. As the task of ranking is global (the rank of every individual depends not only on their own qualifications, but also on every other individuals’ qualifications), ranking is more subtle and vulnerable to manipulation than standard prediction tasks. Towards mitigating unfair discrimination caused by inaccuracies in rankings, we develop two parallel definitions of evidence-based rankings. The first definition relies on a semantic notion of domination-compatibility: if the training data suggest that members of a set S are more qualified (on average) than the members of T, then a ranking that favors T over S (where T dominates S) is blatantly inconsistent with the evidence, and likely to be discriminatory. The definition asks for domination-compatibility, not just for a pair of sets, but rather for every pair of sets from a rich collection C of subpopulations. The second definition aims at precluding even more general forms of discrimination; this notion of evidence-consistency requires that the ranking must be justified on the basis of consistency with the expectations for every set in the collection C. Somewhat surprisingly, while evidence-consistency is a strictly stronger notion than domination-compatibility when the collection C is predefined, the two notions are equivalent when the collection C may depend on the ranking in question.more » « less
-
Worthy, Darrell A. (Ed.)When making decisions involving risk, people may learn about the risk from descriptions or from experience. The description-experience gap refers to the difference in decision patterns driven by this discrepancy in learning format. Across two experiments, we investigated whether learning from description versus experience differentially affects the direction and the magnitude of a context effect in risky decision making. In Study 1 and 2, a computerized game called the Decisions about Risk Task (DART) was used to measure people’s risk-taking tendencies toward hazard stimuli that exploded probabilistically. The rate at which a context hazard caused harm was manipulated, while the rate at which a focal hazard caused harm was held constant. The format by which this information was learned was also manipulated; it was learned primarily by experience or by description. The results revealed that participants’ behavior toward the focal hazard varied depending on what they had learned about the context hazard. Specifically, there were contrast effects in which participants were more likely to choose a risky behavior toward the focal hazard when the harm rate posed by the context hazard was high rather than low. Critically, these contrast effects were of similar strength irrespective of whether the risk information was learned from experience or description. Participants’ verbal assessments of risk likelihood also showed contrast effects, irrespective of learning format. Although risk information about a context hazard in DART does nothing to affect the objective expected value of risky versus safe behaviors toward focal hazards, it did affect participants’ perceptions and behaviors—regardless of whether the information was learned from description or experience. Our findings suggest that context has a broad-based role in how people assess and make decisions about hazards.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
