Algorithmic agents are used in a variety of competitive decision settings, notably in making pricing decisions in contexts that range from online retail to residential home rentals. Business managers, algorithm designers, legal scholars, and regulators alike are all starting to consider the ramifications of "algorithmic collusion." We study the emergent behavior of multi-armed bandit machine learning algorithms used in situations where agents are competing, but they have no information about the strategic interaction they are engaged in. Using a general-form repeated Prisoner's Dilemma game, agents engage in online learning with no prior model of game structure and no knowledge of competitors' states or actions (e.g., no observation of competing prices). We show that these context-free bandits, with no knowledge of opponents' choices or outcomes, still will consistently learn collusive behavior - what we call "naive collusion." We primarily study this system through an analytical model and examine perturbations to the model through simulations. Our findings have several notable implications for regulators. First, calls to limit algorithms from conditioning on competitors' prices are insufficient to prevent algorithmic collusion. This is a direct result of collusion arising even in the naive setting. Second, symmetry in algorithms can increase collusion potential. This highlights a new, simple mechanism for "hub-and-spoke" algorithmic collusion. A central distributor need not imbue its algorithm with supra-competitive tendencies for apparent collusion to arise; it can simply arise by using certain (common) machine learning algorithms. Finally, we highlight that collusive outcomes depend starkly on the specific algorithm being used, and we highlight market and algorithmic conditions under which it will be unknown a priori whether collusion occurs.

This paper proposes techniques to enhance the performance of non-causal models for causal inference using data from randomized experiments. In domains like advertising, customer retention, and precision medicine, non-causal models that predict outcomes under no intervention are often used to score individuals and rank them according to the expected effectiveness of an intervention (e.g, an ad, a retention incentive, a nudge). However, these scores may not perfectly correspond to intervention effects due to the inherent non-causal nature of the models. To address this limitation, we propose causal fine-tuning and effect calibration, two techniques that leverage experimental data to refine the output of non-causal models for different causal tasks, including effect estimation, effect ordering, and effect classification. They are underpinned by two key advantages. First, they can effectively integrate the predictive capabilities of general non-causal models with the requirements of a causal task in a specific context, allowing decision makers to support diverse causal applications with a "foundational" scoring model. Second, through simulations and an empirical example, we demonstrate that they can outperform the alternative of building a causal-effect model from scratch, particularly when the available experimental data is limited and the non-causal scores already capture substantial information about the relative sizes of causal effects. Overall, this research underscores the practical advantages of combining experimental data with non-causal models to support causal applications.

Causal decision making (CDM) at scale has become a routine part of business, and increasingly CDM is based on machine learning algorithms. For example, businesses often target offers, incentives, and recommendations with the goal of affecting consumer behavior. Recently, we have seen an acceleration of research related to CDM and to causal effect estimation (CEE) using machine learned models. This article highlights an important perspective: CDM is not the same as CEE, and counterintuitively, accurate CEE is not necessary for accurate CDM. Our experience is that this is not well understood by practitioners nor by most researchers. Technically, the estimand of interest is different, and this has important implications both for modeling and for the use of statistical models for CDM. We draw on recent research to highlight three of these implications. (1) We should carefully consider the objective function of the causal machine learning, and if possible, we should optimize for accurate "treatment assignment" rather than for accurate effect-size estimation. (2) Confounding does not have the same effect on CDM as it does on CEE. The upshot here is that for supporting CDM it may be just as good to learn with confounded data as with unconfounded data. Finally, (3) causal statistical modeling may not be necessary at all to support CDM, because there may be (and perhaps often is) a proxy target for statistical modeling that can do as well or better. This observation helps to explain at least one broad common CDM practice that seems "wrong" at first blush: the widespread use of non-causal models for targeting interventions. Our perspective is that these observations open up substantial fertile ground for future research. Whether or not you share our perspective completely, we hope we facilitate future research in this area by pointing to related articles from multiple contributing fields.

This study presents a systematic comparison of methods for individual treatment assignment, a general problem that arises in many applications and has received significant attention from economists, computer scientists, and social scientists. We characterize the various methods proposed in the literature into three general approaches: learning models to predict outcomes, learning models to predict causal effects, and learning models to predict optimal treatment assignments. We show analytically that optimizing for outcome or causal-effect prediction is not the same as optimizing for treatment assignments, and thus we should prefer learning models that optimize for treatment assignments. We then compare and contrast the three approaches empirically in the context of choosing, for each user, the best algorithm for playlist generation in order to optimize engagement. This is the first comparison of the different treatment assignment approaches on a real-world application at scale (based on more than half a billion individual treatment assignments). Our results show (i) that applying different algorithms to different users can improve streams substantially compared to deploying the same algorithm for everyone, (ii) that personalized assignments improve substantially with larger data sets, and (iii) that learning models by optimizing treatment assignments rather than outcome or causal-effect predictions can improve treatment assignment performance by more than 28%.

Predictive modeling applications increasingly use data representing people's behavior, opinions, and interactions. Fine-grained behavior data often has different structure from traditional data, being very high-dimensional and sparse. Models built from these data are quite difficult to interpret, since they contain many thousands or even many millions of features. Listing features with large model coefficients is not sufficient, because the model coefficients do not incorporate information on feature presence, which is key when analysing sparse data. In this paper we introduce two alternatives for explaining predictive models by listing important features. We evaluate these alternatives in terms of explanation "bang for the buck,", i.e., how many examples' inferences are explained for a given number of features listed. The bottom line: (i) The proposed alternatives have double the bang-for-the-buck as compared to just listing the high-coefficient features, and (ii) interestingly, although they come from different sources and motivations, the two new alternatives provide strikingly similar rankings of important features.

The Fifteenth National Conference on Artificial Intelligence (AAAI-98) was held in Madison, Wisconsin, on 26-30 July. The following four workshops were held in conjunction with the conference: (1) Case-Based Reasoning Integrations, (2) Learning for Text Categorization, (3) Predicting the Future: AI Approaches to Time-Series Problems, and (4) Software Tools for Developing Agents.

The immense growth of the web has caused the amount of text available online to skyrocket. The AAAI-98 Workshop on Learning for Text Categorization brought together researchers from many of respective areas. A to share their different experiences four workshops were held in conjunction final panel on the synergistic effects of in tackling similar problems. Specifically, several researchers made tasks, no previous workshop soning system, what the significance the point that making use of linguistic attempted to characterize CBR integration of these synergies is, how they can be structure, as well as using stylistic and issues. This nontextual features of documents, can Workshop highlights included panel and the other discussion periods improve categorization performance.

The 1997 AAAI Workshop on AI Approaches to Fraud Detection and Risk Management brought together over 50 researchers and practitioners to discuss problems of fraud detection, computer intrusion detection, and risk scoring. This article presents highlights, including discussions of problematic issues that are common to these application domains, and proposed solutions that apply a variety of AI techniques.

A false negative means that fraud, bad credit, or intrusion passes unnoticed, with potential loss of revenue or security. This workshop focused primarily papers, 10 of which were selected for with the Fourteenth National on what might loosely be termed presentation at the workshop. These Conference on Artificial Intelligence "improper behavior," which includes 10 papers were grouped into 3 categories. However, Glasgow applying classification techniques to were over 50 attendees, with a balanced does discuss the estimation of "inherent fraud and risk problems, including the mix of university and industry risk," which is the bread and butter use of clustering techniques to generate researchers. We sought participants data, highly skewed distributions ("improper Columbia University, and Phillip Chan to discuss and explore common behavior" occurs far less frequently of Florida Institute of Technology).