Author name cluster

Su Jia

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

9 papers

2 author rows

ICML Conference 2025 Conference Paper

Multi-Armed Bandits with Interference: Bridging Causal Inference and Adversarial Bandits

Su Jia
Peter I. Frazier
Nathan Kallus

Experimentation with interference poses a significant challenge in contemporary online platforms. Prior research on experimentation with interference has concentrated on the final output of a policy. Cumulative performance, while equally important, is less well understood. To address this gap, we introduce the problem of Multi-armed Bandits with Interference (MABI), where the learner assigns an arm to each of $N$ experimental units over $T$ rounds. The reward of each unit in each round depends on the treatments of all units, where the interference between two units decays in their distance. The reward functions are chosen by an adversary and may vary arbitrarily over time and across different units. We first show that the optimal expected regret (against the best fixed-arm policy) is $\tilde O(\sqrt T)$, and can be achieved by a switchback policy. However, the regret (as a random variable) for any switchback policy suffers a high variance, since it does not account for $N$. We propose a policy based on a novel clustered randomization scheme, whose regret (i) is optimal in expectation and (ii) admits a high probability bound that vanishes in $N$.

Details

JMLR Journal 2024 Journal Article

Optimal Decision Tree and Adaptive Submodular Ranking with Noisy Outcomes

Su Jia
Fatemeh Navidi
Viswanath Nagarajan
R. Ravi

In pool-based active learning, the learner is given an unlabeled data set and aims to efficiently learn the unknown hypothesis by querying the labels of the data points. This can be formulated as the classical Optimal Decision Tree (ODT) problem: Given a set of tests, a set of hypotheses, and an outcome for each pair of test and hypothesis, our objective is to find a low-cost testing procedure (i.e., decision tree) that identifies the true hypothesis. This optimization problem has been extensively studied under the assumption that each test generates a deterministic outcome. However, in numerous applications, for example, clinical trials, the outcomes may be uncertain, which renders the ideas in the deterministic setting invalid. In this work, we study a fundamental variant of the ODT problem in which some test outcomes are noisy, even in the more general case where the noise is persistent, i.e., repeating a test gives the same noisy output. Our approximation algorithms provide guarantees that are nearly best possible and hold for the general case of a large number of noisy outcomes per test or per hypothesis where the performance degrades continuously with this number. Furthermore, most of our results hold for a more general problem called Adaptive Submodular Ranking with Noise (ASRN). We numerically evaluated our algorithms for identifying toxic chemicals and learning linear classifiers and observed that our algorithms have costs very close to the information-theoretic minimum. [abs] [ pdf ][ bib ] &copy JMLR 2024. ( edit, beta )

PDF Details

ICML Conference 2023 Conference Paper

Short-lived High-volume Bandits

Su Jia
Nishant Oli
Ian Anderson 0005
Paul Duff
Andrew A. Li
R. Ravi 0001

Modern platforms leverage randomized experiments to make informed decisions from a given set of alternatives. As a particularly challenging scenario, these alternatives can potentially have (i) high volume, with thousands of new items being released each hour, and (ii) short lifetime, either due to the contents’ transient nature, or some underlying non-stationarity that impels the learner to treat the same item as non-identical copies across time. We consider a multiplay bandits model. In each round a set of $k=n^\rho$ actions that will be available for $w$ rounds arrives, each of whose mean reward is drawn from a fixed known distribution. The learner selects a multiset of $n$ actions at a time. We propose an $\ell$-Layered Sieve Policy that recursively refines the action space for $\ell\leq w$ times. We show that for any given $\rho>0$, with suitable $\ell$, the policy achieves $\tilde O (n^{-\min \{\rho, \frac 12 (1+\frac 1w)^{-1}\}})$ regret. We also complement this result with an $\Omega (n^{-\min \{\rho, \frac 12\}})$ lower bound. We further validate the effectiveness of our Sieve Policy via numerical simulations and a field experiment in a large content card serving platform.

Details

ICML Conference 2023 Conference Paper

Smooth Non-stationary Bandits

Su Jia
Qian Xie 0005
Nathan Kallus
Peter I. Frazier

In many applications of online decision making, the environment is non-stationary and it is therefore crucial to use bandit algorithms that handle changes. Most existing approaches are designed to protect against non-smooth changes, constrained only by total variation or Lipschitzness over time, where they guarantee $T^{2/3}$ regret. However, in practice environments are often changing smoothly, so such algorithms may incur higher-than-necessary regret in these settings and do not leverage information on the rate of change. In this paper, we study a non-stationary two-arm bandit problem where we assume an arm’s mean reward is a $\beta$-Hölder function over (normalized) time, meaning it is $(\beta-1)$-times Lipschitz-continuously differentiable. We show the first separation between the smooth and non-smooth regimes by presenting a policy with $T^{3/5}$ regret for $\beta=2$. We complement this result by a $T^{\frac{\beta+1}{2\beta+1}}$ lower bound for any integer $\beta\ge 1$, which matches our upper bound for $\beta=2$.

Details

NeurIPS Conference 2022 Conference Paper

Dynamic Pricing with Monotonicity Constraint under Unknown Parametric Demand Model

Su Jia
Andrew Li
R Ravi

We consider the Continuum Bandit problem where the goal is to find the optimal action under an unknown reward function, with an additional monotonicity constraint (or, "markdown" constraint) that requires that the action sequence be non-increasing. This problem faithfully models a natural single-product dynamic pricing problem, called "markdown pricing", where the objective is to adaptively reduce the price over a finite sales horizon to maximize expected revenues. Jia et al '21 and Chen '21 independently showed a tight $T^{3/4}$ regret bound over $T$ rounds under *minimal* assumptions of unimodality and Lipschitzness in the reward (or, "revenue") function. This bound shows that the demand learning in markdown pricing is harder than unconstrained (i. e. , without the monotonicity constraint) pricing under unknown demand which suffers regret only of the order of $T^{2/3}$ under the same assumptions (Kleinberg '04). However, in practice the demand functions are usually assumed to have certain functional forms (e. g. linear or exponential), rendering the demand-learning easier and suggesting lower regret bounds. We investigate two fundamental questions, assuming the underlying demand curve comes from a given parametric family: (1) Can we improve the $T^{3/4}$ regret bound for markdown pricing, under extra assumptions on the functional forms of the demand functions? (2) Is markdown pricing still harder than unconstrained pricing, under these additional assumptions? To answer these, we introduce a concept called markdown dimension that measures the complexity of the parametric family and present tight regret bounds under this framework, thereby completely settling the aforementioned questions.

PDF Details

NeurIPS Conference 2021 Conference Paper

Greedy Approximation Algorithms for Active Sequential Hypothesis Testing

Kyra Gan
Su Jia
Andrew Li

In the problem of \emph{active sequential hypothesis testing} (ASHT), a learner seeks to identify the \emph{true} hypothesis from among a known set of hypotheses. The learner is given a set of actions and knows the random distribution of the outcome of any action under any true hypothesis. Given a target error $\delta>0$, the goal is to sequentially select the fewest number of actions so as to identify the true hypothesis with probability at least $1 - \delta$. Motivated by applications in which the number of hypotheses or actions is massive (e. g. , genomics-based cancer detection), we propose efficient (greedy, in fact) algorithms and provide the first approximation guarantees for ASHT, under two types of adaptivity. Both of our guarantees are independent of the number of actions and logarithmic in the number of hypotheses. We numerically evaluate the performance of our algorithms using both synthetic and real-world DNA mutation data, demonstrating that our algorithms outperform previously proposed heuristic policies by large margins.

PDF Details

NeurIPS Conference 2019 Conference Paper

Optimal Decision Tree with Noisy Outcomes

Su Jia
Viswanath Nagarajan
Fatemeh Navidi
R Ravi

A fundamental task in active learning involves performing a sequence of tests to identify an unknown hypothesis that is drawn from a known distribution. This problem, known as optimal decision tree induction, has been widely studied for decades and the asymptotically best-possible approximation algorithm has been devised for it. We study a generalization where certain test outcomes are noisy, even in the more general case when the noise is persistent, i. e. , repeating the test on the scenario gives the same noisy output, disallowing simple repetition as a way to gain confidence. We design new approximation algorithms for both the non-adaptive setting, where the test sequence must be fixed a-priori, and the adaptive setting where the test sequence depends on the outcomes of prior tests. Previous work in the area assumed at most a constant number of noisy outcomes per test and per scenario and provided approximation ratios that were problem dependent (such as the minimum probability of a hypothesis). Our new approximation algorithms provide guarantees that are nearly best-possible and work for the general case of a large number of noisy outcomes per test or per hypothesis where the performance degrades smoothly with this number. Our results adapt and generalize methods used for submodular ranking and stochastic set cover. We evaluate the performance of our algorithms on two natural applications with noise: toxic chemical identification and active learning of linear classifiers. Despite our logarithmic theoretical approximation guarantees, our methods give solutions with cost very close to the information theoretic minimum, demonstrating the effectiveness of our methods.

PDF Details

AAAI Conference 2017 Conference Paper

Deep Manifold Learning of Symmetric Positive Definite Matrices with Application to Face Recognition

Zhen Dong
Su Jia
Chi Zhang
Mingtao Pei
Yuwei Wu

In this paper, we aim to construct a deep neural network which embeds high dimensional symmetric positive deﬁnite (SPD) matrices into a more discriminative low dimensional SPD manifold. To this end, we develop two types of basic layers: a 2D fully connected layer which reduces the dimensionality of the SPD matrices, and a symmetrically clean layer which achieves non-linear mapping. Speciﬁcally, we extend the classical fully connected layer such that it is suitable for SPD matrices, and we further show that SPD matrices with symmetric pair elements setting zero operations are still symmetric positive deﬁnite. Finally, we complete the construction of the deep neural network for SPD manifold learning by stacking the two layers. Experiments on several face datasets demonstrate the effectiveness of the proposed method.

PDF Details

AAAI Conference 2016 Conference Paper

Face Video Retrieval via Deep Learning of Binary Hash Representations

Zhen Dong
Su Jia
Tianfu Wu
Mingtao Pei

Retrieving faces from large mess of videos is an attractive research topic with wide range of applications. Its challenging problems are large intra-class variations, and tremendous time and space complexity. In this paper, we develop a new deep convolutional neural network (deep CNN) to learn discriminative and compact binary representations of faces for face video retrieval. The network integrates feature extraction and hash learning into a uniﬁed optimization framework for the optimal compatibility of feature extractor and hash functions. In order to better initialize the network, the low-rank discriminative binary hashing is proposed to pre-learn hash functions during the training procedure. Our method achieves excellent performances on two challenging TV-Series datasets.

PDF Details