Search

Algorithmic recourse -- providing recommendations to those affected negatively by the outcome of an algorithmic system on how they can take action and change that outcome -- has gained attention as a means of giving persons agency in their interactions with artificial intelligence (AI) systems. Recent work has shown that even if an AI decision-making classifier is ``fair'' (according to some reasonable criteria), recourse itself may be unfair due to differences in the initial circumstances of individuals, compounding disparities for marginalized populations and requiring them to exert more effort than others. There is a need to define more methods and metrics for evaluating fairness in recourse that span a range of normative views of the world, and specifically those that take into account time. Time is a critical element in recourse because the longer it takes an individual to act, the more the setting may change due to model or data drift. This paper seeks to close this research gap by proposing two notions of fairness in recourse that are in normative alignment with substantive equality of opportunity, and that consider time. The first considers the (often repeated) effort individuals exert per successful recourse event, and the second considers time per successful recourse event. Building upon an agent-based framework for simulating recourse, this paper demonstrates how much effort is needed to overcome disparities in initial circumstances. We then proposes an intervention to improve the fairness of recourse by rewarding effort, and compare it to existing strategies.

A companion paper (Bell, 2001) formulated a model of the cochlea as a surface acoustic wave (SAW) resonator. This supporting paper seeks to give a working account of the sensing elements in the ear and how they operate together to create a SAW resonator. A key feature of any such device is that the interdigital transducers alternate in polarity, an arrangement ideal for launching and detecting surface waves. Translated to the ear, the three rows of outer hair cells (OHCs) are conjectured to be the interdigital transducers. In the simplest (degenerate) SAW resonator, only a single set of three electrodes is required to create resonance between the fingers, a situation presumed to apply in the cochlea, where OHC2 is assumed to respond in antiphase to OHC1 and 3. The antiphasic response is not to displacement, but to intracochlear fluid pressure. An examination of the literature interprets OHCs as responding directly to pressure via their cell bodies, and two populations, with opposite response polarities, are observed. Whether an OHC behaves in one way or the other depends on its membrane potential and turgor pressure, so it is conjectured that OHC1/3 operate at a membrane potential of about –70 mV, whereas in OHC2 it is about –50 mV. At low sound pressure levels, two mechanisms for creating an electrical response in OHCs are identified: one involves the piezoelectric response of the OHC wall to pressure, the other a transient sodium current which acts as a biological ‘transistor’ to amplify the transducer voltage.

In 1857 Helmholtz proposed that the ear contained an array of sympathetic resonators, like piano strings, which served to give the ear its fine frequency discrimination. Since the discovery that most healthy human ears emit faint, pure tones (spontaneous otoacoustic emissions), it has been possible to view these narrowband signals as the continuous ringing of the resonant elements. But what are the elements? It is noteworhty that motile outer hair cells lie in a precise crystal-like array with their sensitive stereocilia in contact with the tectorial membrane, a gelatinous structure with an observed surface tension. This paper therefore speculates that ripples (surface tension waves) on the lower surface of the tectorial membrane propagate to and fro between neighbouring cells. This mechanism defines a surface acoustic wave (SAW) resonator, and relies on the outer hair cells directly sensing intracochlear fluid pressure through their cell bodies; in this way the proposal revisits the resonance theory of hearing. The SAW resonator acts as a regenerative receiver of acoustic energy, a topology which was invoked in 1948 by Gold, who later drew the analogy to an 'underwater piano' to describe the cochlea's problem of how it could vibrate with high Q while immersed in fluid. The proposal also gives a physical description of the cochlear amplifier postulated by Davis in 1983. An active array of resonating cavities driven by outer hair cells can explain spontaneous emissions, the shape of the basilar membrane tuning curve, and evoked emissions, among others, and could relate strongly to music. At levels above which the cochlear amplifier saturates, ripples on the tectorial membrane can still be identified, this time due to vibration of the tectorial membrane against the sharp vestibular lip. This second putative mechanism provides time delays between initiation of the ripple by acoustic pressure variations and its detection by the inner hair cells, and so represents an alternative way of interpreting the traveling wave. Thus, by invoking two ways of generating ripples on the tectorial membrane, a comprehensive account of cochlear mechanics can be constructed, unifying a resonance theory (at low levels) with a traveling wave picture (at high levels).

In 1857 Helmholtz proposed that the ear contained an array of sympathetic resonators, like piano strings, which served to give the ear its fine frequency discrimination. Since the discovery that most healthy human ears emit faint, pure tones (spontaneous otoacoustic emissions), it has been possible to view these narrowband signals as the continuous ringing of the resonant elements. But what are the elements? We note that motile outer hair cells lie in a precise crystal-like array with their sensitive stereocilia in contact with the gelatinous tectorial membrane. This paper therefore proposes that ripples on the surface of the tectorial membrane propagate to and fro between neighbouring cells. The resulting array of active resonators accounts for spontaneous emissions, the shape of the ear’s tuning curve, cochlear echoes, and could relate strongly to music. By identifying the resonating elements that eluded Helmholtz, this hypothesis revives the resonance theory of hearing, displaced this century by the traveling wave picture, and locates the regenerative receiver invoked by Gold in 1948.

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.

This paper sets out a theory of how feedback cooling in the brain switches on consciousness. It explains how cooling reduces thermal noise to the point where macroscale quantum phenomena - crucially Bose-Einstein condensation and long-range coherence - can operate at body temperature. It takes the core idea from Stapp that mind and brain interact via some sort of oscillator and then focuses on a likely candidate: neuronal arrays identified by Stapp as cortical minicolumns. Feedback cooling allows amplifiers to act as refrigerators, and when applied to minicolumns it is suggested that the units perform like quantum accelerators, solid-state devices devised to supercharge standard computers. When the accelerator is idle, as in sleep, we have a neural computer operating unconsciously, but when feedback cooling is activated by thalamocortical loops, it produces a Bose-Einstein condensate, quantum computation, and consciousness. The model explains how macroscale quantum phenomena can operate in a warm and noisy brain, how and why consciousness evolved, and gives insight into puzzling unconscious states like sleepwalking. The model is testable, predicting that cold states in the brain are detectable by magnetic resonance thermometry., Comment: 40 pages, 3 figures. Revised text, additional references

The ``impossibility theorem'' -- which is considered foundational in algorithmic fairness literature -- asserts that there must be trade-offs between common notions of fairness and performance when fitting statistical models, except in two special cases: when the prevalence of the outcome being predicted is equal across groups, or when a perfectly accurate predictor is used. However, theory does not always translate to practice. In this work, we challenge the implications of the impossibility theorem in practical settings. First, we show analytically that, by slightly relaxing the impossibility theorem (to accommodate a \textit{practitioner's} perspective of fairness), it becomes possible to identify a large set of models that satisfy seemingly incompatible fairness constraints. Second, we demonstrate the existence of these models through extensive experiments on five real-world datasets. We conclude by offering tools and guidance for practitioners to understand when -- and to what degree -- fairness along multiple criteria can be achieved. For example, if one allows only a small margin-of-error between metrics, there exists a large set of models simultaneously satisfying \emph{False Negative Rate Parity}, \emph{False Positive Rate Parity}, and \emph{Positive Predictive Value Parity}, even when there is a moderate prevalence difference between groups. This work has an important implication for the community: achieving fairness along multiple metrics for multiple groups (and their intersections) is much more possible than was previously believed., Comment: 14 pages, 3 figures, 1 table

Increasingly, laws are being proposed and passed by governments around the world to regulate Artificial Intelligence (AI) systems implemented into the public and private sectors. Many of these regulations address the transparency of AI systems, and related citizen-aware issues like allowing individuals to have the right to an explanation about how an AI system makes a decision that impacts them. Yet, almost all AI governance documents to date have a significant drawback: they have focused on what to do (or what not to do) with respect to making AI systems transparent, but have left the brunt of the work to technologists to figure out how to build transparent systems. We fill this gap by proposing a novel stakeholder-first approach that assists technologists in designing transparent, regulatory compliant systems. We also describe a real-world case-study that illustrates how this approach can be used in practice.

This article assesses the likely generalizability of educational treatment-effect estimates from regression discontinuity designs (RDDs) when treatment assignment is based on academic pretest scores. Our assessment uses data on outcome and pretest measures from six educational experiments, ranging from preschool through high school, to estimate RDD generalization bias. We then compare those estimates (reported as standardized effect sizes) with the What Works Clearinghouse (WWC) standard for acceptable bias size ([less than or equal] to 0.05[delta]) for two target populations, one spanning a half-standard deviation pretest-score range and another spanning a full-standard deviation pretest-score range. Our results meet this standard for all 18 study/outcome/pretest scenarios examined given the narrower target population, and for 15 scenarios given the broader target population. Fortunately, two of the three exceptions represent pronounced "ceiling effects" that can be identified empirically, making it possible to avoid unwarranted RDD generalizations, and the third exception is very close to the WWC standard. [This is the online version of an article published in "Journal of Research on Educational Effectiveness" (ISSN 1934-5747).]