The problem of uncertainty is a feature of real world robotics problems and any control framework must contend with it in order to succeed in real applications tasks. Reinforcement Learning is no different, and epistemic uncertainty arising from model uncertainty or misspecification is a challenge well captured by the sim-to-real gap. A simple solution to this issue is domain randomization (DR), which unfortunately can result in conservative agents. As a remedy to this conservativeness, the use of universal policies that take additional information about the randomized domain has risen as an alternative solution, along with recurrent neural network-based controllers. Uncertainty-aware universal policies present a particularly compelling solution able to account for system identification uncertainties during deployment. In this paper, we reveal that the challenge of efficiently optimizing uncertainty-aware policies can be fundamentally reframed as solving the convex coverage set (CCS) problem within a multi-objective reinforcement learning (MORL) context. By introducing a novel Markov decision process (MDP) framework where each domain's performance is treated as an independent objective, we unify the training of uncertainty-aware policies with MORL approaches. This connection enables the application of MORL algorithms for domain randomization (DR), allowing for more efficient policy optimization. To illustrate this, we focus on the linear utility function, which aligns with the expectation in DR formulations, and propose a series of algorithms adapted from the MORL literature to solve the CCS, demonstrating their ability to enhance the performance of uncertainty-aware policies.

Traditional network management algorithms have relied on prior knowledge of system models and networking scenarios. In practice, a universal optimization framework is desirable where a sole optimization module can be readily applied to arbitrary network management tasks without any knowledge of the system. To this end, knowledge-free optimization techniques are necessary whose operations are independent of scenario-specific information including objective functions, system parameters, and network setups. The major challenge of this paradigm-shifting approach is the requirement of a hyperintelligent black-box optimizer that can establish efficient decision-making policies using its internal reasoning capabilities. This article presents a novel knowledge-free network management paradigm with the power of foundation models called large language models (LLMs). Trained on vast amounts of datasets, LLMs can understand important contexts from input prompts containing minimal system information, thereby offering remarkable inference performance even for entirely new tasks. Pretrained LLMs can be potentially leveraged as foundation models for versatile network optimization. By eliminating the dependency on prior knowledge, LLMs can be seamlessly applied for various network management tasks. The viability of this approach is demonstrated for resource management problems using GPT-3.5-Turbo. Numerical results validate that knowledge-free LLM optimizers are able to achieve comparable performance to existing knowledge-based optimization algorithms. H. Lee is with the Department of Electrical Engineering and the Artificial Intelligence Graduate School, Ulsan National Institute of Science and Technology (UNIST), Ulsan, 44919, Korea.

We develop a decision making framework to cast the problem of learning a ranking policy for search or recommendation engines in a two-sided e-commerce marketplace as an expected reward optimization problem using observational data. As a value allocation mechanism, the ranking policy allocates retrieved items to the designated slots so as to maximize the user utility from the slotted items, at any given stage of the shopping journey. The objective of this allocation can in turn be defined with respect to the underlying probabilistic user browsing model as the expected number of interaction events on presented items matching the user intent, given the ranking context. Through recognizing the effect of ranking as an intervention action to inform users' interactions with slotted items and the corresponding economic value of the interaction events for the marketplace, we formulate the expected reward of the marketplace as the collective value from all presented ranking actions. The key element in this formulation is a notion of context value distribution, which signifies not only the attribution of value to ranking interventions within a session but also the distribution of marketplace reward across user sessions. We build empirical estimates for the expected reward of the marketplace from observational data that account for the heterogeneity of economic value across session contexts as well as the distribution shifts in learning from observational user activity data. The ranking policy can then be trained by optimizing the empirical expected reward estimates via standard Bayesian inference techniques. We report empirical results for a product search ranking task in a major e-commerce platform demonstrating the fundamental trade-offs governed by ranking polices trained on empirical reward estimates with respect to extreme choices of the context value distribution.

We consider coverless steganography where a Large Language Model (LLM) drives an arithmetic coding decoder to generate stego-texts. An efficient method should embed secret message bits in as few language tokens as possible, while still keeping the stego-text natural and fluent. We show that on the individual token level, this problem is mathematically equivalent to maximizing the entropy of a replacement probability distribution of the next token generation, subject to a constraint on the KL divergence between the chosen probability distribution and the original distribution given by the LLM. A closed-form solution is provided for the optimization problem, which can be computed efficiently. Several important practical issues are also tackled: 1) An often-overlooked tokenization mismatch issue is resolved with a simple prompt selection approach, 2) The combination of the optimized distribution and the vocabulary truncation technique is considered, and 3) The combination of the optimized distribution with other sequence-level selection heuristics to further enhance the efficiency and reliability is studied.

--Connected and autonomous vehicles (CA Vs) have garnered significant attention due to their extended perception range and enhanced sensing coverage. T o address challenges such as blind spots and obstructions, CA Vs employ vehicle-to-vehicle (V2V) communications to aggregate sensory data from surrounding vehicles. However, cooperative perception is often constrained by the limitations of achievable network throughput and channel quality. In this paper, we propose a channel-aware throughput maximization approach to facilitate CA V data fusion, leveraging a self-supervised autoencoder for adaptive data compression. We formulate the problem as a mixed integer programming (MIP) model, which we decompose into two sub-problems to derive optimal data rate and compression ratio solutions under given link conditions. An autoencoder is then trained to minimize bitrate with the determined compression ratio, and a fine-tuning strategy is employed to further reduce spectrum resource consumption. Experimental evaluation on the OpenCOOD platform demonstrates the effectiveness of our proposed algorithm, showing more than 20.19% improvement in network throughput and a 9.38% increase in average precision (AP@IoU) compared to state-of-the-art methods, with an optimal latency of 19.99 ms. Index T erms --Cooperative perception, throughput optimization, connected and autonomous driving (CA V). Recently, autonomous driving has emerged as a promising technology for smart cities. By leveraging communication and artificial intelligence (AI) technologies, autonomous driving can significantly enhance the performance of a city's transportation system. This improvement is achieved through real-time perception of road conditions and precise object detection from onboard sensors (such as radars, LiDARs, and cameras), thereby improving road safety without human intervention [1]. Moreover, the ability of autonomous vehicles to adapt to dynamic environments and communicate with surrounding infrastructure and vehicles is crucial for maintaining the timeliness and accuracy of collected data, thereby enhancing the overall system performance [2]-[9]. Joint perception among connected and autonomous vehicles (CA Vs) is a key enabler to overcome the limitations of individual agent sensing capabilities [10].

Optimization methods are widely employed in deep learning to identify and mitigate undesired model responses. While gradient-based techniques have proven effective for image models, their application to language models is hindered by the discrete nature of the input space. This study introduces a novel optimization approach, termed the \emph{functional homotopy} method, which leverages the functional duality between model training and input generation. By constructing a series of easy-to-hard optimization problems, we iteratively solve these problems using principles derived from established homotopy methods. We apply this approach to jailbreak attack synthesis for large language models (LLMs), achieving a $20\%-30\%$ improvement in success rate over existing methods in circumventing established safe open-source models such as Llama-2 and Llama-3.

We study the problem of minimizing the expectation of smooth nonconvex functions with the help of several parallel workers whose role is to compute stochastic gradients. In particular, we focus on the challenging situation where the workers' compute times are arbitrarily heterogeneous and random. In the simpler regime characterized by arbitrarily heterogeneous but deterministic compute times, Tyurin and Richt\'arik (NeurIPS 2023) recently designed the first theoretically optimal asynchronous SGD method, called Rennala SGD, in terms of a novel complexity notion called time complexity. The starting point of our work is the observation that Rennala SGD can have arbitrarily bad performance in the presence of random compute times -- a setting it was not designed to handle. To advance our understanding of stochastic optimization in this challenging regime, we propose a new asynchronous SGD method, for which we coin the name MindFlayer SGD. Our theory and empirical results demonstrate the superiority of MindFlayer SGD over existing baselines, including Rennala SGD, in cases when the noise is heavy tailed.

The use of preference annotated data for training machine learning models has a long history going back to early algorithms for recommender systems and market research (Bonilla et al., 2010; Boutilier, 2002; Guo and Sanner, 2010). These days preference optimization algorithms are receiving renewed attention since they are a natural candidate for shaping the outputs of deep learning systems, such as large language models (Ouyang et al., 2022; Team et al., 2024) or control policies, via human feedback (Azar et al., 2023; Christiano et al., 2017; Rafailov et al., 2023). Arguably, preference optimization algorithms can also be a natural choice even when direct human feedback is not available but one instead aims to optimize a machine learning model based on feedback from a hand-coded or learned critic function (judging desirability of solutions). Here preference optimization methods are useful since they let us optimize the model to achieve desired outcomes based on relative rankings between outcomes alone (rather than requiring absolute labels or carefully crafted reward functions). Among preference optimization approaches, those based on directly using preference data - as opposed to casting preference optimization as reinforcement learning from (human) feedback - such as DPO (Rafailov et al., 2023), have emerged as particularly successful since they only require access to an offline dataset of paired preference data, and are fairly robust to application domain and hyperparameter settings. However, algorithms within this class make specific assumptions tailored to their application domain. They were designed to optimize LLMs from human feedback in the form of comparisons of generated sentences and thus, by design, require paired preference data (since they directly model a specific choice of preference distribution). We are interested in finding algorithms that are more flexible, and applicable in settings where the assumptions underlying DPO do not apply.

Extrapolation methods use the last few iterates of an optimization algorithm to produce a better estimate of the optimum. They were shown to achieve optimal convergence rates in a deterministic setting using simple gradient iterates. Here, we study extrapolation methods in a stochastic setting, where the iterates are produced by either a simple or an accelerated stochastic gradient algorithm. We first derive convergence bounds for arbitrary, potentially biased perturbations, then produce asymptotic bounds using the ratio between the variance of the noise and the accuracy of the current point. Finally, we apply this acceleration technique to stochastic algorithms such as SGD, SAGA, SVRG and Katyusha in different settings, and show significant performance gains.

Asynchronous-parallel algorithms have the potential to vastly speed up algorithms by eliminating costly synchronization. However, our understanding of these algorithms is limited because the current convergence theory of asynchronous block coordinate descent algorithms is based on somewhat unrealistic assumptions. In particular, the age of the shared optimization variables being used to update blocks is assumed to be independent of the block being updated. Additionally, it is assumed that the updates are applied to randomly chosen blocks. In this paper, we argue that these assumptions either fail to hold or will imply less efficient implementations.