Padhraic Smyth

Gavin Kerrigan, Justin Ley, Padhraic Smyth

Diffusion generative models have recently been applied to domains where the available data can be seen as a discretization of an underlying function, such as audio signals or time series. However, these models operate directly on the discretized data, and there are no semantics in the modeling process that relate the observed data to the underlying functional forms. We generalize diffusion models to operate directly in function space by developing the foundational theory for such models in terms of Gaussian measures on Hilbert spaces. A significant benefit of our function space point of view is that it allows us to explicitly specify the space of functions we are working in, leading us to develop methods for diffusion generative modeling in Sobolev spaces. Our approach allows us to perform both unconditional and conditional generation of function-valued data. We demonstrate our methods on several synthetic and real-world benchmarks.

Aodong Li, Chen Qiu, Marius Kloft et al.

Anomaly detection (AD) plays a crucial role in many safety-critical application domains. The challenge of adapting an anomaly detector to drift in the normal data distribution, especially when no training data is available for the "new normal," has led to the development of zero-shot AD techniques. In this paper, we propose a simple yet effective method called Adaptive Centered Representations (ACR) for zero-shot batch-level AD. Our approach trains off-the-shelf deep anomaly detectors (such as deep SVDD) to adapt to a set of inter-related training data distributions in combination with batch normalization, enabling automatic zero-shot generalization for unseen AD tasks. This simple recipe, batch normalization plus meta-training, is a highly effective and versatile tool. Our theoretical results guarantee the zero-shot generalization for unseen AD tasks; our empirical results demonstrate the first zero-shot AD results for tabular data and outperform existing methods in zero-shot anomaly detection and segmentation on image data from specialized domains. Code is at https://github.com/aodongli/zero-shot-ad-via-batch-norm

Hyungrok Do, Preston Putzel, Axel Martin et al.

Despite recent advances in algorithmic fairness, methodologies for achieving fairness with generalized linear models (GLMs) have yet to be explored in general, despite GLMs being widely used in practice. In this paper we introduce two fairness criteria for GLMs based on equalizing expected outcomes or log-likelihoods. We prove that for GLMs both criteria can be achieved via a convex penalty term based solely on the linear components of the GLM, thus permitting efficient optimization. We also derive theoretical properties for the resulting fair GLM estimator. To empirically demonstrate the efficacy of the proposed fair GLM, we compare it with other well-known fair prediction methods on an extensive set of benchmark datasets for binary classification and regression. In addition, we demonstrate that the fair GLM can generate fair predictions for a range of response variables, other than binary and continuous outcomes.

Shang Wu, Saatvik Kher, Padhraic Smyth

We address the problem of learning to assign prediction tasks to one agent from a set of available human or AI agents. In particular, we focus on the sequential learning of agent expertise and assignment policies where each agent is constrained to handle a fraction of tasks. We provide a general theoretical characterization of this problem in terms of agent capacities, differences in agent expertise, and task context. We then develop a framework of sequential explore-exploit policy-learning algorithms that seek to maximize overall performance. Experimental results over a variety of tabular, image, and text prediction tasks demonstrate systematic gains from our policy-learning algorithms relative to non-contextual baselines across different types of agents, including LLMs and humans.

Alex Boyd, Sam Showalter, Stephan Mandt et al.

In reasoning about sequential events it is natural to pose probabilistic queries such as "when will event A occur next" or "what is the probability of A occurring before B", with applications in areas such as user modeling, medicine, and finance. However, with machine learning shifting towards neural autoregressive models such as RNNs and transformers, probabilistic querying has been largely restricted to simple cases such as next-event prediction. This is in part due to the fact that future querying involves marginalization over large path spaces, which is not straightforward to do efficiently in such models. In this paper we introduce a general typology for predictive queries in neural autoregressive sequence models and show that such queries can be systematically represented by sets of elementary building blocks. We leverage this typology to develop new query estimation methods based on beam search, importance sampling, and hybrids. Across four large-scale sequence datasets from different application domains, as well as for the GPT-2 language model, we demonstrate the ability to make query answering tractable for arbitrary queries in exponentially-large predictive path-spaces, and find clear differences in cost-accuracy tradeoffs between search and sampling methods.

Markelle Kelly, Padhraic Smyth

The deployment of machine learning classifiers in high-stakes domains requires well-calibrated confidence scores for model predictions. In this paper we introduce the notion of variable-based calibration to characterize calibration properties of a model with respect to a variable of interest, generalizing traditional score-based metrics such as expected calibration error (ECE). In particular, we find that models with near-perfect ECE can exhibit significant miscalibration as a function of features of the data. We demonstrate this phenomenon both theoretically and in practice on multiple well-known datasets, and show that it can persist after the application of existing calibration methods. To mitigate this issue, we propose strategies for detection, visualization, and quantification of variable-based calibration error. We then examine the limitations of current score-based calibration methods and explore potential modifications. Finally, we discuss the implications of these findings, emphasizing that an understanding of calibration beyond simple aggregate measures is crucial for endeavors such as fairness and model interpretability.

Catarina G Belem, Markelle Kelly, Mark Steyvers et al.

_Uncertainty expressions_ such as "probably" or "highly unlikely" are pervasive in human language. While prior work has established that there is population-level agreement in terms of how humans quantitatively interpret these expressions, there has been little inquiry into the abilities of language models in the same context. In this paper, we investigate how language models map linguistic expressions of uncertainty to numerical responses. Our approach assesses whether language models can employ theory of mind in this setting: understanding the uncertainty of another agent about a particular statement, independently of the model's own certainty about that statement. We find that 7 out of 10 models are able to map uncertainty expressions to probabilistic responses in a human-like manner. However, we observe systematically different behavior depending on whether a statement is actually true or false. This sensitivity indicates that language models are substantially more susceptible to bias based on their prior knowledge (as compared to humans). These findings raise important questions and have broad implications for human-AI and AI-AI communication.

Aodong Li, Chen Qiu, Marius Kloft et al.

Selecting informative data points for expert feedback can significantly improve the performance of anomaly detection (AD) in various contexts, such as medical diagnostics or fraud detection. In this paper, we determine a set of theoretical conditions under which anomaly scores generalize from labeled queries to unlabeled data. Motivated by these results, we propose a data labeling strategy with optimal data coverage under labeling budget constraints. In addition, we propose a new learning framework for semi-supervised AD. Extensive experiments on image, tabular, and video data sets show that our approach results in state-of-the-art semi-supervised AD performance under labeling budget constraints.

Alex Boyd, Yuxin Chang, Stephan Mandt et al.

Continuous-time event sequences, i.e., sequences consisting of continuous time stamps and associated event types ("marks"), are an important type of sequential data with many applications, e.g., in clinical medicine or user behavior modeling. Since these data are typically modeled autoregressively (e.g., using neural Hawkes processes or their classical counterparts), it is natural to ask questions about future scenarios such as "what kind of event will occur next" or "will an event of type $A$ occur before one of type $B$". Unfortunately, some of these queries are notoriously hard to address since current methods are limited to naive simulation, which can be highly inefficient. This paper introduces a new typology of query types and a framework for addressing them using importance sampling. Example queries include predicting the $n^\text{th}$ event type in a sequence and the hitting time distribution of one or more event types. We also leverage these findings further to be applicable for estimating general "$A$ before $B$" type of queries. We prove theoretically that our estimation method is effectively always better than naive simulation and show empirically based on three real-world datasets that it is on average 1,000 times more efficient than existing approaches.

Eshant English, Eliot Wong-Toi, Matteo Fontana et al.

Conformal prediction provides machine learning models with prediction sets that offer theoretical guarantees, but the underlying assumption of exchangeability limits its applicability to time series data. Furthermore, existing approaches struggle to handle multi-step ahead prediction tasks, where uncertainty estimates across multiple future time points are crucial. We propose JANET (Joint Adaptive predictioN-region Estimation for Time-series), a novel framework for constructing conformal prediction regions that are valid for both univariate and multivariate time series. JANET generalises the inductive conformal framework and efficiently produces joint prediction regions with controlled K-familywise error rates, enabling flexible adaptation to specific application needs. Our empirical evaluation demonstrates JANET's superior performance in multi-step prediction tasks across diverse time series datasets, highlighting its potential for reliable and interpretable uncertainty quantification in sequential data.

Shang Wu, Hongyu Yao, Catarina Belem et al.

Artificial intelligence (AI) is being increasingly integrated into human problem-solving, yet its effects on individual skill development remain unclear. We examine how both AI usage and informativeness can shape learning in the context of a controlled logical reasoning task with on-demand access to AI assistance. We find that greater AI usage is associated with weaker skill development: heavy AI users underperform relative to comparable peers, whereas light AI users perform similarly to matched users who do not use AI. We also find in our study that these patterns are mediated by AI informativeness. Low-information AI neither improves immediate performance nor preserves performance after AI assistance is removed, and is linked to weaker learning overall. On the other hand, high-information AI was found to improve short-run performance without reducing post-AI outcomes on average in our experiments, but with heterogeneous effects. Our findings in general suggest that AI can, depending on context, either complement human skill development by amplifying independent reasoning or can act as a substitute that undermines such reasoning, with the implication that regulating AI access and usage will be important for promoting skill development in the presence of AI assistance.

Gavin Kerrigan, Kai Nelson, Padhraic Smyth

Continuous-time event sequences, in which events occur at irregular intervals, are ubiquitous across a wide range of industrial and scientific domains. The contemporary modeling paradigm is to treat such data as realizations of a temporal point process, and in machine learning it is common to model temporal point processes in an autoregressive fashion using a neural network. While autoregressive models are successful in predicting the time of a single subsequent event, their performance can degrade when forecasting longer horizons due to cascading errors and myopic predictions. We propose EventFlow, a non-autoregressive generative model for temporal point processes. The model builds on the flow matching framework in order to directly learn joint distributions over event times, side-stepping the autoregressive process. EventFlow is simple to implement and achieves a 20%-53% lower forecast error than the nearest baseline on standard TPP benchmarks while simultaneously using fewer model calls at sampling time.

Rachel Longjohn, Shang Wu, Saatvik Kher et al.

It is increasingly important to evaluate how text generation systems based on large language models (LLMs) behave, such as their tendency to produce harmful output or their sensitivity to adversarial inputs. Such evaluations often rely on a curated benchmark set of input prompts provided to the LLM, where the output for each prompt may be assessed in a binary fashion (e.g., harmful/non-harmful or does not leak/leaks sensitive information), and the aggregation of binary scores is used to evaluate the LLM. However, existing approaches to evaluation often neglect statistical uncertainty quantification. With an applied statistics audience in mind, we provide background on LLM text generation and evaluation, and then describe a Bayesian approach for quantifying uncertainty in binary evaluation metrics. We focus in particular on uncertainty that is induced by the probabilistic text generation strategies typically deployed in LLM-based systems. We present two case studies applying this approach: 1) evaluating refusal rates on a benchmark of adversarial inputs designed to elicit harmful responses, and 2) evaluating pairwise preferences of one LLM over another on a benchmark of open-ended interactive dialogue examples. We demonstrate how the Bayesian approach can provide useful uncertainty quantification about the behavior of LLM-based systems.

Gavin Kerrigan, Giosue Migliorini, Padhraic Smyth

We study the geometry of conditional optimal transport (COT) and prove a dynamical formulation which generalizes the Benamou-Brenier Theorem. Equipped with these tools, we propose a simulation-free flow-based method for conditional generative modeling. Our method couples an arbitrary source distribution to a specified target distribution through a triangular COT plan, and a conditional generative model is obtained by approximating the geodesic path of measures induced by this COT plan. Our theory and methods are applicable in infinite-dimensional settings, making them well suited for a wide class of Bayesian inverse problems. Empirically, we demonstrate that our method is competitive on several challenging conditional generation tasks, including an infinite-dimensional inverse problem.

Rachel Longjohn, Markelle Kelly, Sameer Singh et al.

In machine learning research, it is common to evaluate algorithms via their performance on standard benchmark datasets. While a growing body of work establishes guidelines for -- and levies criticisms at -- data and benchmarking practices in machine learning, comparatively less attention has been paid to the data repositories where these datasets are stored, documented, and shared. In this paper, we analyze the landscape of these $\textit{benchmark data repositories}$ and the role they can play in improving benchmarking. This role includes addressing issues with both datasets themselves (e.g., representational harms, construct validity) and the manner in which evaluation is carried out using such datasets (e.g., overemphasis on a few datasets and metrics, lack of reproducibility). To this end, we identify and discuss a set of considerations surrounding the design and use of benchmark data repositories, with a focus on improving benchmarking practices in machine learning.

Kareem Ahmed, Catarina G Belem, Padhraic Smyth et al.

Semantic control entails steering LM generations towards satisfying subtle non-lexical constraints, e.g., toxicity, sentiment, or politeness, attributes that can be captured by a sequence-level verifier. It can thus be viewed as sampling from the LM distribution conditioned on the target attribute, a computationally intractable problem due to the non-decomposable nature of the verifier. Existing approaches to LM control either only deal with syntactic constraints which cannot capture the aforementioned attributes, or rely on sampling to explore the conditional LM distribution, an ineffective estimator for low-probability events. In this work, we leverage a verifier's gradient information to efficiently reason over all generations that satisfy the target attribute, enabling precise steering of LM generations by reweighing the next-token distribution. Starting from an initial sample, we create a local LM distribution favoring semantically similar sentences. This approximation enables the tractable computation of an expected sentence embedding. We use this expected embedding, informed by the verifier's evaluation at the initial sample, to estimate the probability of satisfying the constraint, which directly informs the update to the next-token distribution. We evaluated the effectiveness of our approach in controlling the toxicity, sentiment, and topic-adherence of LMs yielding generations satisfying the constraint with high probability (>95%) without degrading their quality.

Yuxin Chang, Alex Boyd, Cao Xiao et al.

Marked temporal point processes (MTPPs) model sequences of events occurring at irregular time intervals, with wide-ranging applications in fields such as healthcare, finance and social networks. We propose the state-space point process (S2P2) model, a novel and performant model that leverages techniques derived for modern deep state-space models (SSMs) to overcome limitations of existing MTPP models, while simultaneously imbuing strong inductive biases for continuous-time event sequences that other discrete sequence models (i.e., RNNs, transformers) do not capture. Inspired by the classical linear Hawkes processes, we propose an architecture that interleaves stochastic jump differential equations with nonlinearities to create a highly expressive intensity-based MTPP model, without the need for restrictive parametric assumptions for the intensity. Our approach enables efficient training and inference with a parallel scan, bringing linear complexity and sublinear scaling while retaining expressivity to MTPPs. Empirically, S2P2 achieves state-of-the-art predictive likelihoods across eight real-world datasets, delivering an average improvement of 33% over the best existing approaches.

Sam Showalter, Alex Boyd, Padhraic Smyth et al.

Given a pre-trained classifier and multiple human experts, we investigate the task of online classification where model predictions are provided for free but querying humans incurs a cost. In this practical but under-explored setting, oracle ground truth is not available. Instead, the prediction target is defined as the consensus vote of all experts. Given that querying full consensus can be costly, we propose a general framework for online Bayesian consensus estimation, leveraging properties of the multivariate hypergeometric distribution. Based on this framework, we propose a family of methods that dynamically estimate expert consensus from partial feedback by producing a posterior over expert and model beliefs. Analyzing this posterior induces an interpretable trade-off between querying cost and classification performance. We demonstrate the efficacy of our framework against a variety of baselines on CIFAR-10H and ImageNet-16H, two large-scale crowdsourced datasets.

Mark Steyvers, Catarina Belem, Padhraic Smyth

Large language models (LLMs) are increasingly used in decision-making contexts, but when they present answers without signaling low confidence, users may unknowingly act on erroneous outputs. Prior work shows that LLMs maintain internal uncertainty signals, yet their expressed confidence is often miscalibrated and poorly discriminates between correct and incorrect answers. We investigate whether supervised fine-tuning can improve models' ability to communicate uncertainty and whether such improvements generalize across tasks and domains. We fine-tune LLMs on datasets spanning general knowledge, mathematics, and open-ended trivia, and evaluate two metacognitive tasks: (1) single-question confidence estimation, where the model assigns a numeric certainty to its answer, and (2) pairwise confidence comparison, where the model selects which of two answers it is more likely to answer correctly. We assess generalization to unseen domains, including medical and legal reasoning. Results show that fine-tuning improves calibration (alignment between stated confidence and accuracy) and discrimination (higher confidence for correct vs. incorrect responses) within and across domains. However, gains are task-specific: training on single-question calibration does not transfer to pairwise comparison, and vice versa. Multitask fine-tuning yields broader gains, lowering calibration error and strengthening discrimination in out-of-domain evaluations. This suggests that uncertainty communication in LLMs is trainable but requires multitask training to generalize effectively.

Ola Rønning, Eric Nalisnick, Christophe Ley et al.

Stein variational gradient descent (SVGD) [Liu and Wang, 2016] performs approximate Bayesian inference by representing the posterior with a set of particles. However, SVGD suffers from variance collapse, i.e. poor predictions due to underestimating uncertainty [Ba et al., 2021], even for moderately-dimensional models such as small Bayesian neural networks (BNNs). To address this issue, we generalize SVGD by letting each particle parameterize a component distribution in a mixture model. Our method, Stein Mixture Inference (SMI), optimizes a lower bound to the evidence (ELBO) and introduces user-specified guides parameterized by particles. SMI extends the Nonlinear SVGD framework [Wang and Liu, 2019] to the case of variational Bayes. SMI effectively avoids variance collapse, judging by a previously described test developed for this purpose, and performs well on standard data sets. In addition, SMI requires considerably fewer particles than SVGD to accurately estimate uncertainty for small BNNs. The synergistic combination of NSVGD, ELBO optimization and user-specified guides establishes a promising approach towards variational Bayesian inference in the case of tall and wide data.

Coen Adler, Yuxin Chang, Felix Draxler et al.

The recent development of foundation models for time series data has generated considerable interest in using such models across a variety of applications. Although foundation models achieve state-of-the-art predictive performance, their calibration properties remain relatively underexplored, despite the fact that calibration can be critical for many practical applications. In this paper, we investigate the calibration-related properties of five recent time series foundation models and two competitive baselines. We perform a series of systematic evaluations assessing model calibration (i.e., over- or under-confidence), effects of varying prediction heads, and calibration under long-term autoregressive forecasting. We find that time series foundation models are consistently better calibrated than baseline models and tend not to be either systematically over- or under-confident, in contrast to the overconfidence often seen in other deep learning models.

Giosue Migliorini, Padhraic Smyth

Systems of interacting continuous-time Markov chains are a powerful model class, but inference is typically intractable in high dimensional settings. Auxiliary information, such as noisy observations, is typically only available at discrete times, and incorporating it via a Doob's $h-$transform gives rise to an intractable posterior process that requires approximation. We introduce Latent Interacting Particle Systems, a model class parameterizing the generator of each Markov chain in the system. Our inference method involves estimating look-ahead functions (twist potentials) that anticipate future information, for which we introduce an efficient parameterization. We incorporate this approximation in a twisted Sequential Monte Carlo sampling scheme. We demonstrate the effectiveness of our approach on a challenging posterior inference task for a latent SIRS model on a graph, and on a neural model for wildfire spread dynamics trained on real data.

Markelle Kelly, Alex Boyd, Sam Showalter et al.

Applications of machine learning often involve making predictions based on both model outputs and the opinions of human experts. In this context, we investigate the problem of querying experts for class label predictions, using as few human queries as possible, and leveraging the class probability estimates of pre-trained classifiers. We develop a general Bayesian framework for this problem, modeling expert correlation via a joint latent representation, enabling simulation-based inference about the utility of additional expert queries, as well as inference of posterior distributions over unobserved expert labels. We apply our approach to two real-world medical classification problems, as well as to CIFAR-10H and ImageNet-16H, demonstrating substantial reductions relative to baselines in the cost of querying human experts while maintaining high prediction accuracy.

Markelle Kelly, Mohammad Tahaei, Padhraic Smyth et al.

While gender bias in large language models (LLMs) has been extensively studied in many domains, uses of LLMs in e-commerce remain largely unexamined and may reveal novel forms of algorithmic bias and harm. Our work investigates this space, developing data-driven taxonomic categories of gender bias in the context of product description generation, which we situate with respect to existing general purpose harms taxonomies. We illustrate how AI-generated product descriptions can uniquely surface gender biases in ways that require specialized detection and mitigation approaches. Further, we quantitatively analyze issues corresponding to our taxonomic categories in two models used for this task -- GPT-3.5 and an e-commerce-specific LLM -- demonstrating that these forms of bias commonly occur in practice. Our results illuminate unique, under-explored dimensions of gender bias, such as assumptions about clothing size, stereotypical bias in which features of a product are advertised, and differences in the use of persuasive language. These insights contribute to our understanding of three types of AI harms identified by current frameworks: exclusionary norms, stereotyping, and performance disparities, particularly for the context of e-commerce.

Aodong Li, Yunhan Zhao, Chen Qiu et al.

Large language models (LLMs) have shown their potential in long-context understanding and mathematical reasoning. In this paper, we study the problem of using LLMs to detect tabular anomalies and show that pre-trained LLMs are zero-shot batch-level anomaly detectors. That is, without extra distribution-specific model fitting, they can discover hidden outliers in a batch of data, demonstrating their ability to identify low-density data regions. For LLMs that are not well aligned with anomaly detection and frequently output factual errors, we apply simple yet effective data-generating processes to simulate synthetic batch-level anomaly detection datasets and propose an end-to-end fine-tuning strategy to bring out the potential of LLMs in detecting real anomalies. Experiments on a large anomaly detection benchmark (ODDS) showcase i) GPT-4 has on-par performance with the state-of-the-art transductive learning-based anomaly detection methods and ii) the efficacy of our synthetic dataset and fine-tuning strategy in aligning LLMs to this task.

Mark Steyvers, Heliodoro Tejeda, Aakriti Kumar et al.

As artificial intelligence (AI) systems, particularly large language models (LLMs), become increasingly integrated into decision-making processes, the ability to trust their outputs is crucial. To earn human trust, LLMs must be well calibrated such that they can accurately assess and communicate the likelihood of their predictions being correct. Whereas recent work has focused on LLMs' internal confidence, less is understood about how effectively they convey uncertainty to users. Here we explore the calibration gap, which refers to the difference between human confidence in LLM-generated answers and the models' actual confidence, and the discrimination gap, which reflects how well humans and models can distinguish between correct and incorrect answers. Our experiments with multiple-choice and short-answer questions reveal that users tend to overestimate the accuracy of LLM responses when provided with default explanations. Moreover, longer explanations increased user confidence, even when the extra length did not improve answer accuracy. By adjusting LLM explanations to better reflect the models' internal confidence, both the calibration gap and the discrimination gap narrowed, significantly improving user perception of LLM accuracy. These findings underscore the importance of accurate uncertainty communication and highlight the effect of explanation length in influencing user trust in AI-assisted decision-making environments. Code and Data can be found at https://osf.io/y7pr6/ . Journal publication can be found on Nature Machine Intelligence at https://www.nature.com/articles/s42256-024-00976-7 .

Yuxin Chang, Alex Boyd, Padhraic Smyth

Neural marked temporal point processes have been a valuable addition to the existing toolbox of statistical parametric models for continuous-time event data. These models are useful for sequences where each event is associated with a single item (a single type of event or a "mark") -- but such models are not suited for the practical situation where each event is associated with a set of items. In this work, we develop a general framework for modeling set-valued data in continuous-time, compatible with any intensity-based recurrent neural point process model. In addition, we develop inference methods that can use such models to answer probabilistic queries such as "the probability of item $A$ being observed before item $B$," conditioned on sequence history. Computing exact answers for such queries is generally intractable for neural models due to both the continuous-time nature of the problem setting and the combinatorially-large space of potential outcomes for each event. To address this, we develop a class of importance sampling methods for querying with set-based sequences and demonstrate orders-of-magnitude improvements in efficiency over direct sampling via systematic experiments with four real-world datasets. We also illustrate how to use this framework to perform model selection using likelihoods that do not involve one-step-ahead prediction.

Gavin Kerrigan, Giosue Migliorini, Padhraic Smyth

We propose Functional Flow Matching (FFM), a function-space generative model that generalizes the recently-introduced Flow Matching model to operate in infinite-dimensional spaces. Our approach works by first defining a path of probability measures that interpolates between a fixed Gaussian measure and the data distribution, followed by learning a vector field on the underlying space of functions that generates this path of measures. Our method does not rely on likelihoods or simulations, making it well-suited to the function space setting. We provide both a theoretical framework for building such models and an empirical evaluation of our techniques. We demonstrate through experiments on several real-world benchmarks that our proposed FFM method outperforms several recently proposed function-space generative models.

Markelle Kelly, Aakriti Kumar, Padhraic Smyth et al.

Improving our understanding of how humans perceive AI teammates is an important foundation for our general understanding of human-AI teams. Extending relevant work from cognitive science, we propose a framework based on item response theory for modeling these perceptions. We apply this framework to real-world experiments, in which each participant works alongside another person or an AI agent in a question-answering setting, repeatedly assessing their teammate's performance. Using this experimental data, we demonstrate the use of our framework for testing research questions about people's perceptions of both AI agents and other people. We contrast mental models of AI teammates with those of human teammates as we characterize the dimensionality of these mental models, their development over time, and the influence of the participants' own self-perception. Our results indicate that people expect AI agents' performance to be significantly better on average than the performance of other humans, with less variation across different types of problems. We conclude with a discussion of the implications of these findings for human-AI interaction.

Gavin Kerrigan, Padhraic Smyth, Mark Steyvers

An increasingly common use case for machine learning models is augmenting the abilities of human decision makers. For classification tasks where neither the human or model are perfectly accurate, a key step in obtaining high performance is combining their individual predictions in a manner that leverages their relative strengths. In this work, we develop a set of algorithms that combine the probabilistic output of a model with the class-level output of a human. We show theoretically that the accuracy of our combination model is driven not only by the individual human and model accuracies, but also by the model's confidence. Empirical results on image classification with CIFAR-10 and a subset of ImageNet demonstrate that such human-model combinations consistently have higher accuracies than the model or human alone, and that the parameters of the combination method can be estimated effectively with as few as ten labeled datapoints.

Tijl De Bie, Luc De Raedt, José Hernández-Orallo et al.

Given the complexity of typical data science projects and the associated demand for human expertise, automation has the potential to transform the data science process. Key insights: * Automation in data science aims to facilitate and transform the work of data scientists, not to replace them. * Important parts of data science are already being automated, especially in the modeling stages, where techniques such as automated machine learning (AutoML) are gaining traction. * Other aspects are harder to automate, not only because of technological challenges, but because open-ended and context-dependent tasks require human interaction.

Aodong Li, Alex Boyd, Padhraic Smyth et al.

We consider the problem of online learning in the presence of distribution shifts that occur at an unknown rate and of unknown intensity. We derive a new Bayesian online inference approach to simultaneously infer these distribution shifts and adapt the model to the detected changes by integrating ideas from change point detection, switching dynamical systems, and Bayesian online learning. Using a binary 'change variable,' we construct an informative prior such that--if a change is detected--the model partially erases the information of past model updates by tempering to facilitate adaptation to the new data distribution. Furthermore, the approach uses beam search to track multiple change-point hypotheses and selects the most probable one in hindsight. Our proposed method is model-agnostic, applicable in both supervised and unsupervised learning settings, suitable for an environment of concept drifts or covariate drifts, and yields improvements over state-of-the-art Bayesian online learning approaches.

Alex Boyd, Robert Bamler, Stephan Mandt et al.

Continuous-time event data are common in applications such as individual behavior data, financial transactions, and medical health records. Modeling such data can be very challenging, in particular for applications with many different types of events, since it requires a model to predict the event types as well as the time of occurrence. Recurrent neural networks that parameterize time-varying intensity functions are the current state-of-the-art for predictive modeling with such data. These models typically assume that all event sequences come from the same data distribution. However, in many applications event sequences are generated by different sources, or users, and their characteristics can be very different. In this paper, we extend the broad class of neural marked point process models to mixtures of latent embeddings, where each mixture component models the characteristic traits of a given user. Our approach relies on augmenting these models with a latent variable that encodes user characteristics, represented by a mixture model over user behavior that is trained via amortized variational inference. We evaluate our methods on four large real-world datasets and demonstrate systematic improvements from our approach over existing work for a variety of predictive metrics such as log-likelihood, next event ranking, and source-of-sequence identification.

Disi Ji, Padhraic Smyth, Mark Steyvers

We investigate the problem of reliably assessing group fairness when labeled examples are few but unlabeled examples are plentiful. We propose a general Bayesian framework that can augment labeled data with unlabeled data to produce more accurate and lower-variance estimates compared to methods based on labeled data alone. Our approach estimates calibrated scores for unlabeled examples in each group using a hierarchical latent variable model conditioned on labeled examples. This in turn allows for inference of posterior distributions with associated notions of uncertainty for a variety of group fairness metrics. We demonstrate that our approach leads to significant and consistent reductions in estimation error across multiple well-known fairness datasets, sensitive attributes, and predictive models. The results show the benefits of using both unlabeled data and Bayesian inference in terms of assessing whether a prediction model is fair or not.

Disi Ji, Robert L. Logan, Padhraic Smyth et al.

Recent advances in machine learning have led to increased deployment of black-box classifiers across a wide variety of applications. In many such situations there is a critical need to both reliably assess the performance of these pre-trained models and to perform this assessment in a label-efficient manner (given that labels may be scarce and costly to collect). In this paper, we introduce an active Bayesian approach for assessment of classifier performance to satisfy the desiderata of both reliability and label-efficiency. We begin by developing inference strategies to quantify uncertainty for common assessment metrics such as accuracy, misclassification cost, and calibration error. We then propose a general framework for active Bayesian assessment using inferred uncertainty to guide efficient selection of instances for labeling, enabling better performance assessment with fewer labels. We demonstrate significant gains from our proposed active Bayesian approach via a series of systematic empirical experiments assessing the performance of modern neural classifiers (e.g., ResNet and BERT) on several standard image and text classification datasets.

Eric Nalisnick, José Miguel Hernández-Lobato, Padhraic Smyth

Dropout regularization of deep neural networks has been a mysterious yet effective tool to prevent overfitting. Explanations for its success range from the prevention of "co-adapted" weights to it being a form of cheap Bayesian inference. We propose a novel framework for understanding multiplicative noise in neural networks, considering continuous distributions as well as Bernoulli noise (i.e. dropout). We show that multiplicative noise induces structured shrinkage priors on a network's weights. We derive the equivalence through reparametrization properties of scale mixtures and without invoking any approximations. Given the equivalence, we then show that dropout's Monte Carlo training objective approximates marginal MAP estimation. We leverage these insights to propose a novel shrinkage framework for resnets, terming the prior 'automatic depth determination' as it is the natural analog of automatic relevance determination for network depth. Lastly, we investigate two inference strategies that improve upon the aforementioned MAP approximation in regression benchmarks.

Disi Ji, Eric Nalisnick, Padhraic Smyth

Analysis of flow cytometry data is an essential tool for clinical diagnosis of hematological and immunological conditions. Current clinical workflows rely on a manual process called gating to classify cells into their canonical types. This dependence on human annotation limits the rate, reproducibility, and complexity of flow cytometry analysis. In this paper, we propose using Mondrian processes to perform automated gating by incorporating prior information of the kind used by gating technicians. The method segments cells into types via Bayesian nonparametric trees. Examining the posterior over trees allows for interpretable visualizations and uncertainty quantification - two vital qualities for implementation in clinical practice.

Eric Nalisnick, Padhraic Smyth

Informative Bayesian priors are often difficult to elicit, and when this is the case, modelers usually turn to noninformative or objective priors. However, objective priors such as the Jeffreys and reference priors are not tractable to derive for many models of interest. We address this issue by proposing techniques for learning reference prior approximations: we select a parametric family and optimize a black-box lower bound on the reference prior objective to find the member of the family that serves as a good approximation. We experimentally demonstrate the method's effectiveness by recovering Jeffreys priors and learning the Variational Autoencoder's reference prior.

Tracy Holsclaw, Arthur M. Greene, Andrew W. Robertson et al.

Discrete-time hidden Markov models are a broadly useful class of latent-variable models with applications in areas such as speech recognition, bioinformatics, and climate data analysis. It is common in practice to introduce temporal non-homogeneity into such models by making the transition probabilities dependent on time-varying exogenous input variables via a multinomial logistic parametrization. We extend such models to introduce additional non-homogeneity into the emission distribution using a generalized linear model (GLM), with data augmentation for sampling-based inference. However, the presence of the logistic function in the state transition model significantly complicates parameter inference for the overall model, particularly in a Bayesian context. To address this we extend the recently-proposed Polya-Gamma data augmentation approach to handle non-homogeneous hidden Markov models (NHMMs), allowing the development of an efficient Markov chain Monte Carlo (MCMC) sampling scheme. We apply our model and inference scheme to 30 years of daily rainfall in India, leading to a number of insights into rainfall-related phenomena in the region. Our proposed approach allows for fully Bayesian analysis of relatively complex NHMMs on a scale that was not possible with previous methods. Software implementing the methods described in the paper is available via the R package NHMM.

Eric Nalisnick, Padhraic Smyth

We extend Stochastic Gradient Variational Bayes to perform posterior inference for the weights of Stick-Breaking processes. This development allows us to define a Stick-Breaking Variational Autoencoder (SB-VAE), a Bayesian nonparametric version of the variational autoencoder that has a latent representation with stochastic dimensionality. We experimentally demonstrate that the SB-VAE, and a semi-supervised variant, learn highly discriminative latent representations that often outperform the Gaussian VAE's.

Eric Nalisnick, Anima Anandkumar, Padhraic Smyth

Corrupting the input and hidden layers of deep neural networks (DNNs) with multiplicative noise, often drawn from the Bernoulli distribution (or 'dropout'), provides regularization that has significantly contributed to deep learning's success. However, understanding how multiplicative corruptions prevent overfitting has been difficult due to the complexity of a DNN's functional form. In this paper, we show that when a Gaussian prior is placed on a DNN's weights, applying multiplicative noise induces a Gaussian scale mixture, which can be reparameterized to circumvent the problematic likelihood function. Analysis can then proceed by using a type-II maximum likelihood procedure to derive a closed-form expression revealing how regularization evolves as a function of the network's weights. Results show that multiplicative noise forces weights to become either sparse or invariant to rescaling. We find our analysis has implications for model compression as it naturally reveals a weight pruning rule that starkly contrasts with the commonly used signal-to-noise ratio (SNR). While the SNR prunes weights with large variances, seeing them as noisy, our approach recognizes their robustness and retains them. We empirically demonstrate our approach has a strong advantage over the SNR heuristic and is competitive to retraining with soft targets produced from a teacher model.

Kevin Bache, Dennis DeCoste, Padhraic Smyth

We describe a general framework for online adaptation of optimization hyperparameters by `hot swapping' their values during learning. We investigate this approach in the context of adaptive learning rate selection using an explore-exploit strategy from the multi-armed bandit literature. Experiments on a benchmark neural network show that the hot swapping approach leads to consistently better solutions compared to well-known alternatives such as AdaDelta and stochastic gradient with exhaustive hyperparameter search.

James Foulds, Levi Boyles, Christopher Dubois et al.

In the internet era there has been an explosion in the amount of digital text information available, leading to difficulties of scale for traditional inference algorithms for topic models. Recent advances in stochastic variational inference algorithms for latent Dirichlet allocation (LDA) have made it feasible to learn topic models on large-scale corpora, but these methods do not currently take full advantage of the collapsed representation of the model. We propose a stochastic algorithm for collapsed variational Bayesian inference for LDA, which is simpler and more efficient than the state of the art method. We show connections between collapsed variational Bayesian inference and MAP estimation for LDA, and leverage these connections to prove convergence properties of the proposed algorithm. In experiments on large-scale text corpora, the algorithm was found to converge faster and often to a better solution than the previous method. Human-subject experiments also demonstrated that the method can learn coherent topics in seconds on small corpora, facilitating the use of topic models in interactive document analysis software.

Dmitry Y. Pavlov, Heikki Mannila, Padhraic Smyth

Large sparse sets of binary transaction data with millions of records and thousands of attributes occur in various domains: customers purchasing products, users visiting web pages, and documents containing words are just three typical examples. Real-time query selectivity estimation (the problem of estimating the number of rows in the data satisfying a given predicate) is an important practical problem for such databases. We investigate the application of probabilistic models to this problem. In particular, we study a Markov random field (MRF) approach based on frequent sets and maximum entropy, and compare it to the independence model and the Chow-Liu tree model. We find that the MRF model provides substantially more accurate probability estimates than the other methods but is more expensive from a computational and memory viewpoint. To alleviate the computational requirements we show how one can apply bucket elimination and clique tree approaches to take advantage of structure in the models and in the queries. We provide experimental results on two large real-world transaction datasets.

Michal Rosen-Zvi, Thomas Griffiths, Mark Steyvers et al.

We introduce the author-topic model, a generative model for documents that extends Latent Dirichlet Allocation (LDA; Blei, Ng, & Jordan, 2003) to include authorship information. Each author is associated with a multinomial distribution over topics and each topic is associated with a multinomial distribution over words. A document with multiple authors is modeled as a distribution over topics that is a mixture of the distributions associated with the authors. We apply the model to a collection of 1,700 NIPS conference papers and 160,000 CiteSeer abstracts. Exact inference is intractable for these datasets and we use Gibbs sampling to estimate the topic and author distributions. We compare the performance with two other generative models for documents, which are special cases of the author-topic model: LDA (a topic model) and a simple author model in which each author is associated with a distribution over words rather than a distribution over topics. We show topics recovered by the author-topic model, and demonstrate applications to computing similarity between authors and entropy of author output.

Sergey Kirshner, Padhraic Smyth, Andrew Robertson

We consider the problem of modeling discrete-valued vector time series data using extensions of Chow-Liu tree models to capture both dependencies across time and dependencies across variables. Conditional Chow-Liu tree models are introduced, as an extension to standard Chow-Liu trees, for modeling conditional rather than joint densities. We describe learning algorithms for such models and show how they can be used to learn parsimonious representations for the output distributions in hidden Markov models. These models are applied to the important problem of simulating and forecasting daily precipitation occurrence for networks of rain stations. To demonstrate the effectiveness of the models, we compare their performance versus a number of alternatives using historical precipitation data from Southwestern Australia and the Western United States. We illustrate how the structure and parameters of the models can be used to provide an improved meteorological interpretation of such data.

Ian Porteous, Alexander T. Ihler, Padhraic Smyth et al.

Nonparametric Bayesian approaches to clustering, information retrieval, language modeling and object recognition have recently shown great promise as a new paradigm for unsupervised data analysis. Most contributions have focused on the Dirichlet process mixture models or extensions thereof for which efficient Gibbs samplers exist. In this paper we explore Gibbs samplers for infinite complexity mixture models in the stick breaking representation. The advantage of this representation is improved modeling flexibility. For instance, one can design the prior distribution over cluster sizes or couple multiple infinite mixture models (e.g. over time) at the level of their parameters (i.e. the dependent Dirichlet process model). However, Gibbs samplers for infinite mixture models (as recently introduced in the statistics literature) seem to mix poorly over cluster labels. Among others issues, this can have the adverse effect that labels for the same cluster in coupled mixture models are mixed up. We introduce additional moves in these samplers to improve mixing over cluster labels and to bring clusters into correspondence. An application to modeling of storm trajectories is used to illustrate these ideas.

Arthur Asuncion, Max Welling, Padhraic Smyth et al.

Latent Dirichlet analysis, or topic modeling, is a flexible latent variable framework for modeling high-dimensional sparse count data. Various learning algorithms have been developed in recent years, including collapsed Gibbs sampling, variational inference, and maximum a posteriori estimation, and this variety motivates the need for careful empirical comparisons. In this paper, we highlight the close connections between these approaches. We find that the main differences are attributable to the amount of smoothing applied to the counts. When the hyperparameters are optimized, the differences in performance among the algorithms diminish significantly. The ability of these algorithms to achieve solutions of comparable accuracy gives us the freedom to select computationally efficient approaches. Using the insights gained from this comparative study, we show how accurate topic models can be learned in several seconds on text corpora with thousands of documents.