Tongxin Li

Tongxin Li, Yiheng Lin, Shaolei Ren et al.

We study the tradeoff between consistency and robustness in the context of a single-trajectory time-varying Markov Decision Process (MDP) with untrusted machine-learned advice. Our work departs from the typical approach of treating advice as coming from black-box sources by instead considering a setting where additional information about how the advice is generated is available. We prove a first-of-its-kind consistency and robustness tradeoff given Q-value advice under a general MDP model that includes both continuous and discrete state/action spaces. Our results highlight that utilizing Q-value advice enables dynamic pursuit of the better of machine-learned advice and a robust baseline, thus result in near-optimal performance guarantees, which provably improves what can be obtained solely with black-box advice.

Tongxin Li, Ruixiao Yang, Guannan Qu et al.

Machine-learned black-box policies are ubiquitous for nonlinear control problems. Meanwhile, crude model information is often available for these problems from, e.g., linear approximations of nonlinear dynamics. We study the problem of equipping a black-box control policy with model-based advice for nonlinear control on a single trajectory. We first show a general negative result that a naive convex combination of a black-box policy and a linear model-based policy can lead to instability, even if the two policies are both stabilizing. We then propose an adaptive $λ$-confident policy, with a coefficient $λ$ indicating the confidence in a black-box policy, and prove its stability. With bounded nonlinearity, in addition, we show that the adaptive $λ$-confident policy achieves a bounded competitive ratio when a black-box policy is near-optimal. Finally, we propose an online learning approach to implement the adaptive $λ$-confident policy and verify its efficacy in case studies about the CartPole problem and a real-world electric vehicle (EV) charging problem with data bias due to COVID-19.

Jianyi Yang, Pengfei Li, Tongxin Li et al.

This paper studies the problem of Anytime-Competitive Markov Decision Process (A-CMDP). Existing works on Constrained Markov Decision Processes (CMDPs) aim to optimize the expected reward while constraining the expected cost over random dynamics, but the cost in a specific episode can still be unsatisfactorily high. In contrast, the goal of A-CMDP is to optimize the expected reward while guaranteeing a bounded cost in each round of any episode against a policy prior. We propose a new algorithm, called Anytime-Competitive Reinforcement Learning (ACRL), which provably guarantees the anytime cost constraints. The regret analysis shows the policy asymptotically matches the optimal reward achievable under the anytime competitive constraints. Experiments on the application of carbon-intelligent computing verify the reward performance and cost constraint guarantee of ACRL.

Tongxin Li, Chenxi Sun

We tackle the challenge of learning to charge Electric Vehicles (EVs) with Out-of-Distribution (OOD) data. Traditional scheduling algorithms typically fail to balance near-optimal average performance with worst-case guarantees, particularly with OOD data. Model Predictive Control (MPC) is often too conservative and data-independent, whereas Reinforcement Learning (RL) tends to be overly aggressive and fully trusts the data, hindering their ability to consistently achieve the best-of-both-worlds. To bridge this gap, we introduce a novel OOD-aware scheduling algorithm, denoted OOD-Charging. This algorithm employs a dynamic "awareness radius", which updates in real-time based on the Temporal Difference (TD)-error that reflects the severity of OOD. The OOD-Charging algorithm allows for a more effective balance between consistency and robustness in EV charging schedules, thereby significantly enhancing adaptability and efficiency in real-world charging environments. Our results demonstrate that this approach improves the scheduling reward reliably under real OOD scenarios with remarkable shifts of EV charging behaviors caused by COVID-19 in the Caltech ACN-Data.

Fanzeng Xia, Hao Liu, Yisong Yue et al.

In-Context Reinforcement Learning (ICRL) is a frontier paradigm to solve Reinforcement Learning (RL) problems in the foundation model era. While ICRL capabilities have been demonstrated in transformers through task-specific training, the potential of Large Language Models (LLMs) out-of-the-box remains largely unexplored. This paper investigates whether LLMs can generalize cross-domain to perform ICRL under the problem of Dueling Bandits (DB), a stateless preference-based RL setting. We find that the top-performing LLMs exhibit a notable zero-shot capacity for relative decision-making, which translates to low short-term weak regret across all DB environment instances by quickly including the best arm in duels. However, an optimality gap still exists between LLMs and classic DB algorithms in terms of strong regret. LLMs struggle to converge and consistently exploit even when explicitly prompted to do so, and are sensitive to prompt variations. To bridge this gap, we propose an agentic flow framework: LLM with Enhanced Algorithmic Dueling (LEAD), which integrates off-the-shelf DB algorithm support with LLM agents through fine-grained adaptive interplay. We show that LEAD has theoretical guarantees inherited from classic DB algorithms on both weak and strong regret. We validate its efficacy and robustness even with noisy and adversarial prompts. The design of such an agentic framework sheds light on how to enhance the trustworthiness of general-purpose LLMs generalized to in-context decision-making tasks.

Pei Wang, Yu Ding, Mingshen Sun et al.

The big data industry is facing new challenges as concerns about privacy leakage soar. One of the remedies to privacy breach incidents is to encapsulate computations over sensitive data within hardware-assisted Trusted Execution Environments (TEE). Such TEE-powered software is called secure enclaves. Secure enclaves hold various advantages against competing for privacy-preserving computation solutions. However, enclaves are much more challenging to build compared with ordinary software. The reason is that the development of TEE software must follow a restrictive programming model to make effective use of strong memory encryption and segregation enforced by hardware. These constraints transitively apply to all third-party dependencies of the software. If these dependencies do not officially support TEE hardware, TEE developers have to spend additional engineering effort in porting them. High development and maintenance cost is one of the major obstacles against adopting TEE-based privacy protection solutions in production. In this paper, we present our experience and achievements with regard to constructing and continuously maintaining a third-party library supply chain for TEE developers. In particular, we port a large collection of Rust third-party libraries into Intel SGX, one of the most mature trusted computing platforms. Our supply chain accepts upstream patches in a timely manner with SGX-specific security auditing. We have been able to maintain the SGX ports of 159 open-source Rust libraries with reasonable operational costs. Our work can effectively reduce the engineering cost of developing SGX enclaves for privacy-preserving data processing and exchange.

Xudong Wang, Tongxin Li, Chris Ding et al.

Graph data often exhibits complex geometric heterogeneity, where structures with varying local curvature, such as tree-like hierarchies and dense communities, coexist within a single network. Existing geometric GNNs, which embed graphs into single fixed-curvature manifolds or discrete product spaces, struggle to capture this diversity. We introduce Adaptive Riemannian Graph Neural Networks (ARGNN), a novel framework that learns a continuous and anisotropic Riemannian metric tensor field over the graph. It allows each node to determine its optimal local geometry, enabling the model to fluidly adapt to the graph's structural landscape. Our core innovation is an efficient parameterization of the node-wise metric tensor, specializing to a learnable diagonal form that captures directional geometric information while maintaining computational tractability. To ensure geometric regularity and stable training, we integrate a Ricci flow-inspired regularization that smooths the learned manifold. Theoretically, we establish the rigorous geometric evolution convergence guarantee for ARGNN and provide a continuous generalization that unifies prior fixed or mixed-curvature GNNs. Empirically, our method demonstrates superior performance on both homophilic and heterophilic benchmark datasets with the ability to capture diverse structures adaptively. Moreover, the learned geometries both offer interpretable insights into the underlying graph structure and empirically corroborate our theoretical analysis.

Chenbei Lu, Zaiwei Chen, Tongxin Li et al.

Traditional reinforcement learning (RL) assumes the agents make decisions based on Markov decision processes (MDPs) with one-step transition models. In many real-world applications, such as energy management and stock investment, agents can access multi-step predictions of future states, which provide additional advantages for decision making. However, multi-step predictions are inherently high-dimensional: naively embedding these predictions into an MDP leads to an exponential blow-up in state space and the curse of dimensionality. Moreover, existing RL theory provides few tools to analyze prediction-augmented MDPs, as it typically works on one-step transition kernels and cannot accommodate multi-step predictions with errors or partial action-coverage. We address these challenges with three key innovations: First, we propose the \emph{Bayesian value function} to characterize the optimal prediction-aware policy tractably. Second, we develop a novel \emph{Bellman-Jensen Gap} analysis on the Bayesian value function, which enables characterizing the value of imperfect predictions. Third, we introduce BOLA (Bayesian Offline Learning with Online Adaptation), a two-stage model-based RL algorithm that separates offline Bayesian value learning from lightweight online adaptation to real-time predictions. We prove that BOLA remains sample-efficient even under imperfect predictions. We validate our theory and algorithm on synthetic MDPs and a real-world wind energy storage control problem.

Yingrui Zhuang, Lin Cheng, Yuji Cao et al.

Price signals from distribution networks (DNs) guide energy communities (ECs) to adjust energy usage, enabling effective coordination for reliable power system operation. However, this coordination faces significant challenges due to the limited availability of information (i.e., only the aggregated energy usage of ECs is available to DNs), and the high computational burden of accounting for uncertainties and the associated risks through numerous scenarios. To address these challenges, we propose a quantum learning and estimation approach to enhance coordination between DNs and ECs. Specifically, leveraging advanced quantum properties such as quantum superposition and entanglement, we develop a hybrid quantum temporal convolutional network-long short-term memory (Q-TCN-LSTM) model to establish an end-to-end mapping between ECs' responses and the price incentives from DNs. Moreover, we develop a quantum estimation method based on quantum amplitude estimation (QAE) and two phase-rotation circuits to significantly accelerate the optimization process under numerous uncertainty scenarios. Numerical experiments demonstrate that, compared to classical neural networks, the proposed Q-TCN-LSTM model improves the mapping accuracy by 69.2% while reducing the model size by 99.75% and the computation time by 93.9%. Compared to classical Monte Carlo simulation, QAE achieves comparable accuracy with a dramatic reduction in computational time (up to 99.99%) and requires significantly fewer computational resources.

Fanzeng Xia, Yidong Luo, Tinko Sebastian Bartels et al.

Recent research has highlighted that Large Language Models (LLMs), even when trained to generate extended long reasoning steps, still face significant challenges on hard reasoning problems. However, much of the existing literature relies on direct prompting with simple in-context learning examples for evaluation, which largely overlooks advanced techniques to elicit LLMs' deliberate reasoning before drawing conclusions that LLMs hit a performance ceiling. In this paper, we systematically explore the combined potential of in-context search and test-time scaling on super hard reasoning tasks. We find that by employing advanced in-context search prompting to LLMs augmented with internal scaling, one can achieve transformative performance breakthroughs on tasks previously deemed "unsolvable" (e.g., reported success rates below 5%). We provide both empirical results and theoretical analysis of how this combination can unleash LLM reasoning capabilities: i) Empirically, on controlled NP-hard tasks and complex real-world planning benchmarks, our approach achieves up to a 30x improvement in success rates compared to previously reported results without any external mechanisms; ii) Theoretically, we show that in-context search prompting, when combined with internal scaling, significantly extends the complexity class of solvable reasoning problems. These findings challenge prevailing assumptions about the limitations of LLMs on complex tasks, indicating that current evaluation paradigms systematically underestimate their true potential. Our work calls for a critical reassessment of how LLM reasoning is benchmarked and a more robust evaluation strategy that fully captures the true capabilities of contemporary LLMs, which can lead to a better understanding of their operational reasoning boundaries in real-world deployments.

Kaixin Zhang, Hongzhi Wang, Han Hu et al.

Recently, deep learning has been an area of intense research. However, as a kind of computing-intensive task, deep learning highly relies on the scale of GPU memory, which is usually prohibitive and scarce. Although some extensive works have been proposed for dynamic GPU memory management, they are hard to apply to systems with multiple dynamic workloads, such as in-database machine learning systems. In this paper, we demonstrated TENSILE, a method of managing GPU memory in tensor granularity to reduce the GPU memory peak, considering the multiple dynamic workloads. TENSILE tackled the cold-starting and across-iteration scheduling problem existing in previous works. We implemented TENSILE on a deep learning framework built by ourselves and evaluated its performance. The experiment results show that TENSILE can save more GPU memory with less extra overhead than prior works in single and multiple dynamic workloads scenarios.

Anqi Liu, Hao Liu, Tongxin Li et al.

Discovering the complete set of causal relations among a group of variables is a challenging unsupervised learning problem. Often, this challenge is compounded by the fact that there are latent or hidden confounders. When only observational data is available, the problem is ill-posed, i.e. the causal relationships are non-identifiable unless strong modeling assumptions are made. When interventions are available, we provide guarantees on identifiability and learnability under mild assumptions. We assume a linear structural equation model (SEM) with independent latent factors and directed acyclic graph (DAG) relationships among the observables. Since the latent variable inference is based on independent component analysis (ICA), we call this model SEM-ICA. We use the method of moments principle to establish model identifiability. We develop efficient algorithms based on coupled tensor decomposition with linear constraints to obtain scalable and guaranteed solutions. Thus, we provide a principled approach to tackling the joint problem of causal discovery and latent variable inference.

Huibo Wang, Mingshen Sun, Qian Feng et al.

Intel SGX Guard eXtensions (SGX), a hardware-supported trusted execution environment (TEE), is designed to protect security-sensitive applications. However, since enclave applications are developed with memory unsafe languages such as C/C++, traditional memory corruption is not eliminated in SGX. Rust-SGX is the first toolkit providing enclave developers with a memory-language. However, Rust is considered a Systems language and has become the right choice for concurrent applications and web browsers. Many application domains such as Big Data, Machine Learning, Robotics, Computer Vision are more commonly developed in the python programming language. Therefore, Python application developers cannot benefit from secure enclaves like Intel SGX and rust-SGX. To fill this gap, we propose Python-SGX, which is a memory-safe SGX SDK providing enclave developers a memory-safe Python development environment. The key idea is to enable memory-safe Python language in SGX by solving the following key challenges: (1) defining a memory-safe Python interpreter (2)replacing unsafe elements of Python interpreter with safe ones,(3) achieving comparable performance to non-enclave Python applications, and (4) not introducing any unsafe new code or libraries into SGX. We propose to build Python-SGX with PyPy, a Python interpreter written by RPython, which is a subset of Python, and tame unsafe parts in PyPy by formal verification, security hardening, and memory safe language. We have implemented python-SGX and tested it with a series of benchmarks programs. Our evaluation results show that Python-SGX does not cause significant overhead.

Luyi Xing, Xiaolong Bai, Tongxin Li et al.

On modern operating systems, applications under the same user are separated from each other, for the purpose of protecting them against malware and compromised programs. Given the complexity of today's OSes, less clear is whether such isolation is effective against different kind of cross-app resource access attacks (called XARA in our research). To better understand the problem, on the less-studied Apple platforms, we conducted a systematic security analysis on MAC OS~X and iOS. Our research leads to the discovery of a series of high-impact security weaknesses, which enable a sandboxed malicious app, approved by the Apple Stores, to gain unauthorized access to other apps' sensitive data. More specifically, we found that the inter-app interaction services, including the keychain, WebSocket and NSConnection on OS~X and URL Scheme on the MAC OS and iOS, can all be exploited by the malware to steal such confidential information as the passwords for iCloud, email and bank, and the secret token of Evernote. Further, the design of the app sandbox on OS~X was found to be vulnerable, exposing an app's private directory to the sandboxed malware that hijacks its Apple Bundle ID. As a result, sensitive user data, like the notes and user contacts under Evernote and photos under WeChat, have all been disclosed. Fundamentally, these problems are caused by the lack of app-to-app and app-to-OS authentications. To better understand their impacts, we developed a scanner that automatically analyzes the binaries of MAC OS and iOS apps to determine whether proper protection is missing in their code. Running it on hundreds of binaries, we confirmed the pervasiveness of the weaknesses among high-impact Apple apps. Since the issues may not be easily fixed, we built a simple program that detects exploit attempts on OS~X, helping protect vulnerable apps before the problems can be fully addressed.