Jeffrey Chan

AAAI Conference 2026 Conference Paper

Hallucinate Less by Thinking More: Aspect-Based Causal Abstention for Large Language Models

• Vy Nguyen
• Ziqi Xu
• Jeffrey Chan
• Estrid He
• Feng Xia
• Xiuzhen Zhang

Large Language Models (LLMs) often produce fluent but factually incorrect responses, a phenomenon known as hallucination. Abstention, where the model chooses not to answer and instead outputs phrases such as "I don't know", is a common safeguard. However, existing abstention methods typically rely on post-generation signals, such as generation variations or feedback, which limits their ability to prevent unreliable responses in advance. In this paper, we introduce Aspect-Based Causal Abstention (ABCA), a new framework that enables early abstention by analysing the internal diversity of LLM knowledge through causal inference. This diversity reflects the multifaceted nature of parametric knowledge acquired from various sources, representing diverse aspects such as disciplines, legal contexts, or temporal frames. ABCA estimates causal effects conditioned on these aspects to assess the reliability of knowledge relevant to a given query. Based on these estimates, we enable two types of abstention: Type-1, where aspect effects are inconsistent (knowledge conflict), and Type-2, where aspect effects consistently support abstention (knowledge insufficiency). Experiments on standard benchmarks demonstrate that ABCA improves abstention reliability, achieves state-of-the-art performance, and enhances the interpretability of abstention decisions.

AAAI Conference 2025 Conference Paper

BOIDS: High-Dimensional Bayesian Optimization via Incumbent-Guided Direction Lines and Subspace Embeddings

• Lam Ngo
• Huong Ha
• Jeffrey Chan
• Hongyu Zhang

When it comes to expensive black-box optimization problems, Bayesian Optimization (BO) is a well-known and powerful solution. Many real-world applications involve a large number of dimensions, hence scaling BO to high dimension is of much interest. However, state-of-the-art high-dimensional BO methods still suffer from the curse of dimensionality, highlighting the need for further improvements. In this work, we introduce BOIDS, a novel high-dimensional BO algorithm that guides optimization by a sequence of one-dimensional direction lines using a novel tailored line-based optimization procedure. To improve the efficiency, we also propose an adaptive selection technique to identify most optimal lines for each round of line-based optimization. Additionally, we incorporate a subspace embedding technique for better scaling to high-dimensional spaces. We further provide theoretical analysis of our proposed method to analyze its convergence property. Our extensive experimental results show that BOIDS outperforms state-of-the-art baselines on various synthetic and real-world benchmark problems.

TIST Journal 2025 Journal Article

Counterfactual Explanations via Locally-guided Sequential Algorithmic Recourse

• Edward A. Small
• Jeffrey N. Clark
• Christopher J. McWilliams
• Michael Ambler
• Kacper Sokol
• Jeffrey Chan
• Flora D. Salim
• Raul Santos-Rodriguez

Counterfactuals operationalised through algorithmic recourse have become a powerful tool to make artificial intelligence systems explainable. Conceptually, given an individual classified as \(y\) – the factual – we seek actions such that their prediction becomes the desired class \(y^{\prime}\) – the counterfactual. This process offers algorithmic recourse that is (1) easy to customise and interpret, and (2) directly aligned with the goals of each individual. However, the properties of a “good” counterfactual are still largely debated; it remains an open challenge to locate an effective counterfactual along with its corresponding recourse. Some strategies use gradient-driven methods, but these offer no guarantees on the feasibility of the recourse and are open to adversarial attacks on carefully created manifolds. This can lead to unfairness and lack of robustness. Other methods are data-driven, which mostly addresses the feasibility problem at the expense of privacy, security, and secrecy as they require access to the entire training data set. Here, we introduce a model-agnostic technique that composes feasible and actionable counterfactual explanations using locally-acquired information at each step of the algorithmic recourse. Our explainer preserves the privacy of users by only leveraging data that it specifically requires to construct actionable algorithmic recourse, and protects the model by offering transparency solely in the regions deemed necessary for the intervention.

NeurIPS Conference 2025 Conference Paper

MOBO-OSD: Batch Multi-Objective Bayesian Optimization via Orthogonal Search Directions

• Lam Ngo
• Huong Ha
• Jeffrey Chan
• Hongyu Zhang

Bayesian Optimization (BO) is a powerful tool for optimizing expensive black-box objective functions. While extensive research has been conducted on the single-objective optimization problem, the multi-objective optimization problem remains challenging. In this paper, we propose MOBO-OSD, a multi-objective Bayesian Optimization algorithm designed to generate a diverse set of Pareto optimal solutions by solving multiple constrained optimization problems, referred to as MOBO-OSD subproblems, along orthogonal search directions (OSDs) defined with respect to an approximated convex hull of individual objective minima. By employing a well-distributed set of OSDs, MOBO-OSD ensures broad coverage of the objective space, enhancing both solution diversity and hypervolume performance. To further improve the density of the set of the Pareto optimal candidate solutions without requiring an excessive number of subproblems, we leverage a Pareto Front Estimation technique to generate additional solutions in the neighborhood of existing solutions. Additionally, MOBO-OSD supports batch optimization, enabling parallel function evaluations to accelerate the optimization process when resources are available. Through extensive experiments and analysis on a variety of synthetic and real-world benchmark functions with two to six objectives, we demonstrate that MOBO-OSD consistently outperform the state-of-the-art algorithms.

ECAI Conference 2025 Conference Paper

MOCA-HESP: Meta High-Dimensional Bayesian Optimization for Combinatorial and Mixed Spaces via Hyper-Ellipsoid Partitioning

• Lam Ngo
• Huong Ha 0001
• Jeffrey Chan
• Hongyu Zhang 0002

High-dimensional Bayesian Optimization (BO) has attracted significant attention in recent research. However, existing methods have mainly focused on optimizing in continuous domains, while combinatorial (ordinal and categorical) and mixed domains still remain challenging. In this paper, we first propose MOCA-HESP, a novel high-dimensional BO method for combinatorial and mixed variables. The key idea is to leverage the hyper-ellipsoid space partitioning (HESP) technique with different categorical encoders to work with high-dimensional, combinatorial and mixed spaces, while adaptively selecting the optimal encoders for HESP using a multi-armed bandit technique. Our method, MOCA-HESP, is designed as a meta-algorithm such that it can incorporate other combinatorial and mixed BO optimizers to further enhance the optimizers’ performance. Finally, we develop three practical BO methods by integrating MOCA-HESP with state-of-the-art BO optimizers for combinatorial and mixed variables: standard BO, CASMOPOLITAN, and Bounce. Our experimental results on various synthetic and real-world benchmarks show that our methods outperform existing baselines. Our code implementation can be found at https: //github. com/LamNgo1/moca-hesp.

TMLR Journal 2024 Journal Article

High-dimensional Bayesian Optimization via Covariance Matrix Adaptation Strategy

• Lam Ngo
• Huong Ha
• Jeffrey Chan
• Vu Nguyen
• Hongyu Zhang

Bayesian Optimization (BO) is an effective method for finding the global optimum of expensive black-box functions. However, it is well known that applying BO to high-dimensional optimization problems is challenging. To address this issue, a promising solution is to use a local search strategy that partitions the search domain into local regions with high likelihood of containing the global optimum, and then use BO to optimize the objective function within these regions. In this paper, we propose a novel technique for defining the local regions using the Covariance Matrix Adaptation (CMA) strategy. Specifically, we use CMA to learn a search distribution that can estimate the probabilities of data points being the global optimum of the objective function. Based on this search distribution, we then define the local regions consisting of data points with high probabilities of being the global optimum. Our approach serves as a meta-algorithm as it can incorporate existing black-box BO optimizers, such as BO, TuRBO, and BAxUS, to find the global optimum of the objective function within our derived local regions. We evaluate our proposed method on various benchmark synthetic and real-world problems. The results demonstrate that our method outperforms existing state-of-the-art techniques.

EAAI Journal 2023 Journal Article

Automatic meter error detection with a data-driven approach

• Ruimin Chu
• Li Chik
• Jeffrey Chan
• Kurt Gutzmann
• Xiaodong Li

Meter error is one of the main contributing factors to unexpected fuel losses or gains in storage tanks at service stations. Although fuel dispensers are expected to be calibrated to standard accuracy periodically to ensure fair and reliable trade in the fuel market, some fuel retailers are unable to keep up with the standards. The current industry practice relies on onsite inspection to identify the issue, which leads to a cost burden if inspections are scheduled too frequently. To the best of our knowledge, there is no previously reported research tailored to the remote meter error detection problem. In this paper, we propose a novel framework for remote and automatic meter error detection via a data-driven approach based on inventory data and fuel transaction data. Specifically, we propose to use mean shift change point detection methods, including statistical-based as well as deep learning-based methods (LSTM-VAE, VAE, Kernel learning), to approach the problem. We present results on our data sets containing both real-world and simulated meter error data, and further evaluate these methods on several widely-used benchmark datasets, to assess their validity, advantages and limitations. The obtained results show that LSTM-VAE outperforms other models in most of the settings for the meter error dataset and the benchmark datasets.

AAAI Conference 2022 Conference Paper

A Divide and Conquer Algorithm for Predict+Optimize with Non-convex Problems

• Ali Ugur Guler
• Emir Demirović
• Jeffrey Chan
• James Bailey
• Christopher Leckie
• Peter J. Stuckey

The predict+optimize problem combines machine learning and combinatorial optimization by predicting the problem coefficients first and then using these coefficients to solve the optimization problem. While this problem can be solved in two separate stages, recent research shows end to end models can achieve better results. This requires differentiating through a discrete combinatorial function. Models that use differentiable surrogates are prone to approximation errors, while existing exact models are limited to dynamic programming, or they do not generalize well with scarce data. In this work we propose a novel divide and conquer algorithm based on transition points to reason over exact optimization problems and predict the coefficients using the optimization loss. Moreover, our model is not limited to dynamic programming problems. We also introduce a greedy version, which achieves similar results with less computation. In comparison with other predict+optimize frameworks, we show our method outperforms existing exact frameworks and can reason over hard combinatorial problems better than surrogate methods.

JMLR Journal 2022 Journal Article

MurTree: Optimal Decision Trees via Dynamic Programming and Search

• Emir Demirović
• Anna Lukina
• Emmanuel Hebrard
• Jeffrey Chan
• James Bailey
• Christopher Leckie
• Kotagiri Ramamohanarao
• Peter J. Stuckey

Decision tree learning is a widely used approach in machine learning, favoured in applications that require concise and interpretable models. Heuristic methods are traditionally used to quickly produce models with reasonably high accuracy. A commonly criticised point, however, is that the resulting trees may not necessarily be the best representation of the data in terms of accuracy and size. In recent years, this motivated the development of optimal classification tree algorithms that globally optimise the decision tree in contrast to heuristic methods that perform a sequence of locally optimal decisions. We follow this line of work and provide a novel algorithm for learning optimal classification trees based on dynamic programming and search. Our algorithm supports constraints on the depth of the tree and number of nodes. The success of our approach is attributed to a series of specialised techniques that exploit properties unique to classification trees. Whereas algorithms for optimal classification trees have traditionally been plagued by high runtimes and limited scalability, we show in a detailed experimental study that our approach uses only a fraction of the time required by the state-of-the-art and can handle datasets with tens of thousands of instances, providing several orders of magnitude improvements and notably contributing towards the practical use of optimal decision trees. [abs] [ pdf ][ bib ] [ code ] &copy JMLR 2022. ( edit, beta )

AAAI Conference 2020 Conference Paper

Dynamic Programming for Predict+Optimise

• Emir Demirovi?
• Peter J. Stuckey
• Tias Guns
• James Bailey
• Christopher Leckie
• Kotagiri Ramamohanarao
• Jeffrey Chan

We study the predict+optimise problem, where machine learning and combinatorial optimisation must interact to achieve a common goal. These problems are important when optimisation needs to be performed on input parameters that are not fully observed but must instead be estimated using machine learning. We provide a novel learning technique for predict+optimise to directly reason about the underlying combinatorial optimisation problem, offering a meaningful integration of machine learning and optimisation. This is done by representing the combinatorial problem as a piecewise linear function parameterised by the coefficients of the learning model and then iteratively performing coordinate descent on the learning coefficients. Our approach is applicable to linear learning functions and any optimisation problem solvable by dynamic programming. We illustrate the effectiveness of our approach on benchmarks from the literature.

AAAI Conference 2020 Conference Paper

Revisiting Probability Distribution Assumptions for Information Theoretic Feature Selection

• Yuan Sun
• Wei Wang
• Michael Kirley
• Xiaodong Li
• Jeffrey Chan

Feature selection has been shown to be beneficial for many data mining and machine learning tasks, especially for big data analytics. Mutual Information (MI) is a well-known information-theoretic approach used to evaluate the relevance of feature subsets and class labels. However, estimating highdimensional MI poses significant challenges. Consequently, a great deal of research has focused on using low-order MI approximations or computing a lower bound on MI called Variational Information (VI). These methods often require certain assumptions made on the probability distributions of features such that these distributions are realistic yet tractable to compute. In this paper, we reveal two sets of distribution assumptions underlying many MI and VI based methods: Feature Independence Distribution and Geometric Mean Distribution. We systematically analyze their strengths and weaknesses and propose a logical extension called Arithmetic Mean Distribution, which leads to an unbiased and normalised estimation of probability densities. We conduct detailed empirical studies across a suite of 29 real-world classification problems and illustrate improved prediction accuracy of our methods based on the identification of more informative features, thus providing support for our theoretical findings.

IJCAI Conference 2019 Conference Paper

Predict+Optimise with Ranking Objectives: Exhaustively Learning Linear Functions

• Emir Demirovic
• Peter J. Stuckey
• James Bailey
• Jeffrey Chan
• Christopher Leckie
• Kotagiri Ramamohanarao
• Tias Guns

We study the predict+optimise problem, where machine learning and combinatorial optimisation must interact to achieve a common goal. These problems are important when optimisation needs to be performed on input parameters that are not fully observed but must instead be estimated using machine learning. Our contributions are two-fold: 1) we provide theoretical insight into the properties and computational complexity of predict+optimise problems in general, and 2) develop a novel framework that, in contrast to related work, guarantees to compute the optimal parameters for a linear learning function given any ranking optimisation problem. We illustrate the applicability of our framework for the particular case of the unit-weighted knapsack predict+optimise problem and evaluate on benchmarks from the literature.

NeurIPS Conference 2018 Conference Paper

A Likelihood-Free Inference Framework for Population Genetic Data using Exchangeable Neural Networks

• Jeffrey Chan
• Valerio Perrone
• Jeffrey Spence
• Paul Jenkins
• Sara Mathieson
• Yun Song

An explosion of high-throughput DNA sequencing in the past decade has led to a surge of interest in population-scale inference with whole-genome data. Recent work in population genetics has centered on designing inference methods for relatively simple model classes, and few scalable general-purpose inference techniques exist for more realistic, complex models. To achieve this, two inferential challenges need to be addressed: (1) population data are exchangeable, calling for methods that efficiently exploit the symmetries of the data, and (2) computing likelihoods is intractable as it requires integrating over a set of correlated, extremely high-dimensional latent variables. These challenges are traditionally tackled by likelihood-free methods that use scientific simulators to generate datasets and reduce them to hand-designed, permutation-invariant summary statistics, often leading to inaccurate inference. In this work, we develop an exchangeable neural network that performs summary statistic-free, likelihood-free inference. Our framework can be applied in a black-box fashion across a variety of simulation-based tasks, both within and outside biology. We demonstrate the power of our approach on the recombination hotspot testing problem, outperforming the state-of-the-art.

AAMAS Conference 2017 Conference Paper

The Importance of Modelling Realistic Human Behaviour When Planning Evacuation Schedules

• Chaminda Bulumulla
• Lin Padgham
• Dhirendra Singh
• Jeffrey Chan

AAAI Conference 2016 Conference Paper

MOOCs Meet Measurement Theory: A Topic-Modelling Approach

• Jiazhen He
• Benjamin Rubinstein
• James Bailey
• Rui Zhang
• Sandra Milligan
• Jeffrey Chan

This paper adapts topic models to the psychometric testing of MOOC students based on their online forum postings. Measurement theory from education and psychology provides statistical models for quantifying a person’s attainment of intangible attributes such as attitudes, abilities or intelligence. Such models infer latent skill levels by relating them to individuals’ observed responses on a series of items such as quiz questions. The set of items can be used to measure a latent skill if individuals’ responses on them conform to a Guttman scale. Such well-scaled items differentiate between individuals and inferred levels span the entire range from most basic to the advanced. In practice, education researchers manually devise items (quiz questions) while optimising well-scaled conformance. Due to the costly nature and expert requirements of this process, psychometric testing has found limited use in everyday teaching. We aim to develop usable measurement models for highly-instrumented MOOC delivery platforms, by using participation in automatically-extracted online forum topics as items. The challenge is to formalise the Guttman scale educational constraint and incorporate it into topic models. To favour topics that automatically conform to a Guttman scale, we introduce a novel regularisation into non-negative matrix factorisation-based topic modelling. We demonstrate the suitability of our approach with both quantitative experiments on three Coursera MOOCs, and with a qualitative survey of topic interpretability on two MOOCs by domain expert interviews.

ICAPS Conference 2016 Conference Paper

Towards Next Generation Touring: Personalized Group Tours

• Kwan Hui Lim 0001
• Jeffrey Chan
• Christopher Leckie
• Shanika Karunasekera

Recommending and planning tour itineraries are challenging and time-consuming for tourists, hence they may seek tour operators for help. Traditionally tour operators have offered standard tour packages of popular locations, but these packages may not cater to tourist's interests. In addition, tourists may want to travel in a group, e. g. , extended family, and want an operator to help them. We introduce the novel problem of group tour recommendation (GroupTourRec), which involves many challenges: forming tour groups whose members have similar interests; recommending Points-of-Interests (POI) that form the tour itinerary and cater for the group's interests; and assigning guides to lead these tours. For each challenge, we propose solutions involving: clustering for tourist groupings; optimizing a variant of the Orienteering problem for POI recommendations; and integer programming for tour guide assignments. Using a Flickr dataset of seven cities, we compare our proposed approaches against various baselines and observe significant improvements in terms of interest similarity, total/maximum/minimum tour interests and total tour guide expertise.

IJCAI Conference 2015 Conference Paper

Personalized Tour Recommendation Based on User Interests and Points of Interest Visit Durations

• Kwan Hui Lim
• Jeffrey Chan
• Christopher Leckie
• Shanika Karunasekera

Tour recommendation and itinerary planning are challenging tasks for tourists, due to their need to select Points of Interest (POI) to visit in unfamiliar cities, and to select POIs that align with their interest preferences and trip constraints. We propose an algorithm called PERSTOUR for recommending personalized tours using POI popularity and user interest preferences, which are automatically derived from real-life travel sequences based on geotagged photos. Our tour recommendation problem is modelled using a formulation of the Orienteering problem, and considers user trip constraints such as time limits and the need to start and end at specific POIs. In our work, we also reflect levels of user interest based on visit durations, and demonstrate how POI visit duration can be personalized using this time-based user interest. Using a Flickr dataset of four cities, our experiments show the effectiveness of PERSTOUR against various baselines, in terms of tour popularity, interest, recall, precision and F1-score. In particular, our results show the merits of using time-based user interest and personalized POI visit durations, compared to the current practice of using frequency-based user interest and average visit durations.

AAAI Conference 2014 Conference Paper

Reconsidering Mutual Information Based Feature Selection: A Statistical Significance View

• Nguyen Vinh
• Jeffrey Chan
• James Bailey

Multi-Instance Multi-Label (MIML) is a learning framework where an example is associated with multiple labels and represented by a set of feature vectors (multiple instances). In the formalization of MIML learning, instances come from a single source (single view). To leverage multiple information sources (multi-view), we develop a multi-view MIML framework based on hierarchical Bayesian Network, and derive an effective learning algorithm based on variational inference. The model can naturally deal with examples in which some views could be absent (partial examples). On multi-view datasets, it is shown that our method is better than other multi-view and single-view approaches particularly in the presence of partial examples. On single-view benchmarks, extensive evaluation shows that our method is highly competitive or better than other MIML approaches on labeling examples and instances. Moreover, our method can effectively handle datasets with a large number of labels.

IJCAI Conference 2011 Conference Paper

Increasing the Scalability of the Fitting of Generalised Block Models for Social Networks

• Jeffrey Chan
• Samantha Lam
• Conor Hayes

In recent years, the summarisation and decomposition of social networks has become increasingly popular, from community finding to role equivalence. However, these approaches concentrate on one type of model only. Generalised block modelling decomposes a network into independent, interpretable, labeled blocks, where the block labels summarise the relationship between two sets of users. Existing algorithms for fitting generalised block models do not scale beyond networks of 100 vertices. In this paper, we introduce two new algorithms, one based on genetic algorithms and the other on simulated annealing, that is at least two orders of magnitude faster than existing algorithms and obtaining similar accuracy. Using synthetic and real datasets, we demonstrate their efficiency and accuracy and show how generalised block modelling and our new approaches enable tractable network summarisation and modelling of medium sized networks.