For infinite action contextual bandits, smoothed regret and reduction to regression results in state-of-the-art online performance with computational cost independent of the action set: unfortunately, the resulting data exhaust does not have well-defined importance-weights. This frustrates the execution of downstream data science processes such as offline model selection. In this paper we describe an online algorithm with an equivalent smoothed regret guarantee, but which generates well-defined importance weights: in exchange, the online computational cost increases, but only to order smoothness (i.e., still independent of the action set). This removes a key obstacle to adoption of smoothed regret in production scenarios.
Nonstationary phenomena, such as satiation effects in recommendations, have mostly been modeled using bandits with finitely many arms. However, the richer action space provided by linear bandits is often preferred in practice. In this work, we introduce a novel nonstationary linear bandit model, where current rewards are influenced by the learner's past actions in a fixed-size window. Our model, which recovers stationary linear bandits as a special case, leverages two parameters: the window size $m \ge 0$, and an exponent $\gamma$ that captures the rotting ($\gamma < 0)$ or rising ($\gamma > 0$) nature of the phenomenon. When both $m$ and $\gamma$ are known, we propose and analyze a variant of OFUL which minimizes regret against cycling policies. By choosing the cycle length so as to trade-off approximation and estimation errors, we then prove a bound of order $\sqrt{d}\,(m+1)^{\frac{1}{2}+\max\{\gamma,0\}}\,T^{3/4}$ (ignoring log factors) on the regret against the optimal sequence of actions, where $T$ is the horizon and $d$ is the dimension of the linear action space. Through a bandit model selection approach, our results are extended to the case where $m$ and $\gamma$ are unknown. Finally, we complement our theoretical results with experiments against natural baselines.
Stochastic Rising Bandits (SRBs) model sequential decision-making problems in which the expected reward of the available options increases every time they are selected. This setting captures a wide range of scenarios in which the available options are learning entities whose performance improves (in expectation) over time (e.g., online best model selection). While previous works addressed the regret minimization problem, this paper focuses on the fixed-budget Best Arm Identification (BAI) problem for SRBs. In this scenario, given a fixed budget of rounds, we are asked to provide a recommendation about the best option at the end of the identification process. We propose two algorithms to tackle the above-mentioned setting, namely R-UCBE, which resorts to a UCB-like approach, and R-SR, which employs a successive reject procedure. Then, we prove that, with a sufficiently large budget, they provide guarantees on the probability of properly identifying the optimal option at the end of the learning process and on the simple regret. Furthermore, we derive a lower bound on the error probability, matched by our R-SR (up to constants), and illustrate how the need for a sufficiently large budget is unavoidable in the SRB setting. Finally, we numerically validate the proposed algorithms in both synthetic and realistic environments.
Most research on fair machine learning has prioritized optimizing criteria such as Demographic Parity and Equalized Odds. Despite these efforts, there remains a limited understanding of how different bias mitigation strategies affect individual predictions and whether they introduce arbitrariness into the debiasing process. This paper addresses these gaps by exploring whether models that achieve comparable fairness and accuracy metrics impact the same individuals and mitigate bias in a consistent manner. We introduce the FRAME (FaiRness Arbitrariness and Multiplicity Evaluation) framework, which evaluates bias mitigation through five dimensions: Impact Size (how many people were affected), Change Direction (positive versus negative changes), Decision Rates (impact on models' acceptance rates), Affected Subpopulations (who was affected), and Neglected Subpopulations (where unfairness persists). This framework is intended to help practitioners understand the impacts of debiasing processes and make better-informed decisions regarding model selection. Applying FRAME to various bias mitigation approaches across key datasets allows us to exhibit significant differences in the behaviors of debiasing methods. These findings highlight the limitations of current fairness criteria and the inherent arbitrariness in the debiasing process.
Modern NLP systems exhibit a range of biases, which a growing literature on model debiasing attempts to correct. However current progress is hampered by a plurality of definitions of bias, means of quantification, and oftentimes vague relation between debiasing algorithms and theoretical measures of bias. This paper seeks to clarify the current situation and plot a course for meaningful progress in fair learning, with two key contributions: (1) making clear inter-relations among the current gamut of methods, and their relation to fairness theory; and (2) addressing the practical problem of model selection, which involves a trade-off between fairness and accuracy and has led to systemic issues in fairness research. Putting them together, we make several recommendations to help shape future work.
How is it that humans can solve complex planning tasks so efficiently despite limited cognitive resources? One reason is its ability to know how to use its limited computational resources to make clever choices. We postulate that people learn this ability from trial and error (metacognitive reinforcement learning). Here, we systematize models of the underlying learning mechanisms and enhance them with more sophisticated additional mechanisms. We fit the resulting 86 models to human data collected in previous experiments where different phenomena of metacognitive learning were demonstrated and performed Bayesian model selection. Our results suggest that a gradient ascent through the space of cognitive strategies can explain most of the observed qualitative phenomena, and is therefore a promising candidate for explaining the mechanism underlying metacognitive learning.
Linear model trees are regression trees that incorporate linear models in the leaf nodes. This preserves the intuitive interpretation of decision trees and at the same time enables them to better capture linear relationships, which is hard for standard decision trees. But most existing methods for fitting linear model trees are time consuming and therefore not scalable to large data sets. In addition, they are more prone to overfitting and extrapolation issues than standard regression trees. In this paper we introduce PILOT, a new algorithm for linear model trees that is fast, regularized, stable and interpretable. PILOT trains in a greedy fashion like classic regression trees, but incorporates an $L^2$ boosting approach and a model selection rule for fitting linear models in the nodes. The abbreviation PILOT stands for $PI$ecewise $L$inear $O$rganic $T$ree, where `organic' refers to the fact that no pruning is carried out. PILOT has the same low time and space complexity as CART without its pruning. An empirical study indicates that PILOT tends to outperform standard decision trees and other linear model trees on a variety of data sets. Moreover, we prove its consistency in an additive model setting under weak assumptions. When the data is generated by a linear model, the convergence rate is polynomial.
Continual learning seeks to empower models to progressively acquire information from a sequence of tasks. This approach is crucial for many real-world systems, which are dynamic and evolve over time. Recent research has witnessed a surge in the exploration of Graph Neural Networks (GNN) in Node-wise Graph Continual Learning (NGCL), a practical yet challenging paradigm involving the continual training of a GNN on node-related tasks. Despite recent advancements in continual learning strategies for GNNs in NGCL, a thorough theoretical understanding, especially regarding its learnability, is lacking. Learnability concerns the existence of a learning algorithm that can produce a good candidate model from the hypothesis/weight space, which is crucial for model selection in NGCL development. This paper introduces the first theoretical exploration of the learnability of GNN in NGCL, revealing that learnability is heavily influenced by structural shifts due to the interconnected nature of graph data. Specifically, GNNs may not be viable for NGCL under significant structural changes, emphasizing the need to manage structural shifts. To mitigate the impact of structural shifts, we propose a novel experience replay method termed Structure-Evolution-Aware Experience Replay (SEA-ER). SEA-ER features an innovative experience selection strategy that capitalizes on the topological awareness of GNNs, alongside a unique replay strategy that employs structural alignment to effectively counter catastrophic forgetting and diminish the impact of structural shifts on GNNs in NGCL. Our extensive experiments validate our theoretical insights and the effectiveness of SEA-ER.
Feature selection in clustering is a hard task which involves simultaneously the discovery of relevant clusters as well as relevant variables with respect to these clusters. While feature selection algorithms are often model-based through optimised model selection or strong assumptions on the data distribution, we introduce a discriminative clustering model trying to maximise a geometry-aware generalisation of the mutual information called GEMINI with a simple l1 penalty: the Sparse GEMINI. This algorithm avoids the burden of combinatorial feature subset exploration and is easily scalable to high-dimensional data and large amounts of samples while only designing a discriminative clustering model. We demonstrate the performances of Sparse GEMINI on synthetic datasets and large-scale datasets. Our results show that Sparse GEMINI is a competitive algorithm and has the ability to select relevant subsets of variables with respect to the clustering without using relevance criteria or prior hypotheses.
We extend the theory from Fan and Li (2001) on penalized likelihood-based estimation and model-selection to statistical and econometric models which allow for non-negativity constraints on some or all of the parameters, as well as time-series dependence. It differs from classic non-penalized likelihood estimation, where limiting distributions of likelihood-based estimators and test-statistics are non-standard, and depend on the unknown number of parameters on the boundary of the parameter space. Specifically, we establish that the joint model selection and estimation, results in standard asymptotic Gaussian distributed estimators. The results are applied to the rich class of autoregressive conditional heteroskedastic (ARCH) models for the modelling of time-varying volatility. We find from simulations that the penalized estimation and model-selection works surprisingly well even for a large number of parameters. A simple empirical illustration for stock-market returns data confirms the ability of the penalized estimation to select ARCH models which fit nicely the autocorrelation function, as well as confirms the stylized fact of long-memory in financial time series data.
Personalized treatment effect estimates are often of interest in high-stakes applications -- thus, before deploying a model estimating such effects in practice, one needs to be sure that the best candidate from the ever-growing machine learning toolbox for this task was chosen. Unfortunately, due to the absence of counterfactual information in practice, it is usually not possible to rely on standard validation metrics for doing so, leading to a well-known model selection dilemma in the treatment effect estimation literature. While some solutions have recently been investigated, systematic understanding of the strengths and weaknesses of different model selection criteria is still lacking. In this paper, instead of attempting to declare a global `winner', we therefore empirically investigate success- and failure modes of different selection criteria. We highlight that there is a complex interplay between selection strategies, candidate estimators and the data used for comparing them, and provide interesting insights into the relative (dis)advantages of different criteria alongside desiderata for the design of further illuminating empirical studies in this context.
Imitation of expert behaviour is a highly desirable and safe approach to the problem of sequential decision making. We provide an easy-to-implement, novel algorithm for imitation learning under a strict data paradigm, in which the agent must learn solely from data collected a priori. This paradigm allows our algorithm to be used for environments in which safety or cost are of critical concern. Our algorithm requires no additional hyper-parameter tuning beyond any standard batch reinforcement learning (RL) algorithm, making it an ideal baseline for such data-strict regimes. Furthermore, we provide formal sample complexity guarantees for the algorithm in finite Markov Decision Problems. In doing so, we formally demonstrate an unproven claim from Kearns & Singh (1998). On the empirical side, our contribution is twofold. First, we develop a practical, robust and principled evaluation protocol for offline RL methods, making use of only the dataset provided for model selection. This stands in contrast to the vast majority of previous works in offline RL, which tune hyperparameters on the evaluation environment, limiting the practical applicability when deployed in new, cost-critical environments. As such, we establish precedent for the development and fair evaluation of offline RL algorithms. Second, we evaluate our own algorithm on challenging continuous control benchmarks, demonstrating its practical applicability and competitiveness with state-of-the-art performance, despite being a simpler algorithm.
Machine learning models play a vital role in time series forecasting. These models, however, often overlook an important element: point uncertainty estimates. Incorporating these estimates is crucial for effective risk management, informed model selection, and decision-making.To address this issue, our research introduces a method for uncertainty estimation. We employ a surrogate Gaussian process regression model. It enhances any base regression model with reasonable uncertainty estimates. This approach stands out for its computational efficiency. It only necessitates training one supplementary surrogate and avoids any data-specific assumptions. Furthermore, this method for work requires only the presence of the base model as a black box and its respective training data. The effectiveness of our approach is supported by experimental results. Using various time-series forecasting data, we found that our surrogate model-based technique delivers significantly more accurate confidence intervals. These techniques outperform both bootstrap-based and built-in methods in a medium-data regime. This superiority holds across a range of base model types, including a linear regression, ARIMA, gradient boosting and a neural network.
In this paper, we build on the analysis of Muni Toke and Yoshida (2020) and conduct several empirical studies using high-frequency financial data. Muni Toke and Yoshida (2020) showed the consistency and asymptotic behavior of the Cox-type model estimators for relative intensities of orders in the limit order book, and then by using high-frequency trading data for 36 stocks traded on the Paris Stock Exchange, they carry out model selection and trading sign prediction. In this study, we add new covariates and carry out model selection and trading sign prediction using high-frequency trading data for 222 stocks traded on the Tokyo Stock Exchange. We not only show that the Cox-type model performs well in the Japanese market as well as in the Euronext Paris market, but also present the key factors for more accurate estimation. We also suggest how often the covariates should be calibrated.
The explosion of data and its ever increasing complexity in the last few years, has made MLOps systems more prone to failure, and new tools need to be embedded in such systems to avoid such failure. In this demo, we will introduce crucial tools in the observability module of a MLOps system that target difficult issues like data drfit and model version control for optimum model selection. We believe integrating these features in our MLOps pipeline would go a long way in building a robust system immune to early stage ML system failures.
As predictive models -- e.g., from machine learning -- give likely outcomes, they may be used to reason on the effect of an intervention, a causal-inference task. The increasing complexity of health data has opened the door to a plethora of models, but also the Pandora box of model selection: which of these models yield the most valid causal estimates? Here we highlight that classic machine-learning model selection does not select the best outcome models for causal inference. Indeed, causal model selection should control both outcome errors for each individual, treated or not treated, whereas only one outcome is observed. Theoretically, simple risks used in machine learning do not control causal effects when treated and non-treated population differ too much. More elaborate risks build proxies of the causal error using ``nuisance'' re-weighting to compute it on the observed data. But does computing these nuisance adds noise to model selection? Drawing from an extensive empirical study, we outline a good causal model-selection procedure: using the so-called $R\text{-risk}$; using flexible estimators to compute the nuisance models on the train set; and splitting out 10\% of the data to compute risks.
Offline model selection (OMS), that is, choosing the best policy from a set of many policies given only logged data, is crucial for applying offline RL in real-world settings. One idea that has been extensively explored is to select policies based on the mean squared Bellman error (MSBE) of the associated Q-functions. However, previous work has struggled to obtain adequate OMS performance with Bellman errors, leading many researchers to abandon the idea. To this end, we elucidate why previous work has seen pessimistic results with Bellman errors and identify conditions under which OMS algorithms based on Bellman errors will perform well. Moreover, we develop a new estimator of the MSBE that is more accurate than prior methods. Our estimator obtains impressive OMS performance on diverse discrete control tasks, including Atari games.
The {\it straight-through estimator} (STE) is commonly used to optimize quantized neural networks, yet its contexts of effective performance are still unclear despite empirical successes.To make a step forward in this comprehension, we apply STE to a well-understood problem: {\it sparse support recovery}. We introduce the {\it Support Exploration Algorithm} (SEA), a novel algorithm promoting sparsity, and we analyze its performance in support recovery (a.k.a. model selection) problems. SEA explores more supports than the state-of-the-art, leading to superior performance in experiments, especially when the columns of $A$ are strongly coherent.The theoretical analysis considers recovery guarantees when the linear measurements matrix $A$ satisfies the {\it Restricted Isometry Property} (RIP).The sufficient conditions of recovery are comparable but more stringent than those of the state-of-the-art in sparse support recovery. Their significance lies mainly in their applicability to an instance of the STE.
The surprising ability of Large Language Models (LLMs) to perform well on complex reasoning with only few-shot chain-of-thought prompts is believed to emerge only in very large-scale models (100+ billion parameters). We show that such abilities can, in fact, be distilled down from GPT-3.5 ($\ge$ 175B) to T5 variants ($\le$ 11B). We propose model specialization, to specialize the model's ability towards a target task. The hypothesis is that large models (commonly viewed as larger than 100B) have strong modeling power, but are spread on a large spectrum of tasks. Small models (commonly viewed as smaller than 10B) have limited model capacity, but if we concentrate their capacity on a specific target task, the model can achieve a decent improved performance. We use multi-step math reasoning as our testbed because it is a very typical emergent ability. We show two important aspects of model abilities: (1). there exists a very complex balance/ tradeoff between language models' multi-dimensional abilities; (2). by paying the price of decreased generic ability, we can clearly lift up the scaling curve of models smaller than 10B towards a specialized multi-step math reasoning ability. We further give comprehensive discussions about important design choices for better generalization, including the tuning data format, the start model checkpoint, and a new model selection method. We hope our practice and discoveries can serve as an important attempt towards specialized smaller models in the new research paradigm set by LLMs.
Bayesian model comparison (BMC) offers a principled approach for assessing the relative merits of competing computational models and propagating uncertainty into model selection decisions. However, BMC is often intractable for the popular class of hierarchical models due to their high-dimensional nested parameter structure. To address this intractability, we propose a deep learning method for performing BMC on any set of hierarchical models which can be instantiated as probabilistic programs. Since our method enables amortized inference, it allows efficient re-estimation of posterior model probabilities and fast performance validation prior to any real-data application. In a series of extensive validation studies, we benchmark the performance of our method against the state-of-the-art bridge sampling method and demonstrate excellent amortized inference across all BMC settings. We then showcase our method by comparing four hierarchical evidence accumulation models that have previously been deemed intractable for BMC due to partly implicit likelihoods. Additionally, we demonstrate how transfer learning can be leveraged to enhance training efficiency. We provide reproducible code for all analyses and an open-source implementation of our method.