Search

In this work, we introduce the Learnable Response Scoring Function (LARS) for Uncertainty Estimation (UE) in generative Large Language Models (LLMs). Current scoring functions for probability-based UE, such as length-normalized scoring and semantic contribution-based weighting, are designed to solve specific aspects of the problem but exhibit limitations, including the inability to handle biased probabilities and under-performance in low-resource languages like Turkish. To address these issues, we propose LARS, a scoring function that leverages supervised data to capture complex dependencies between tokens and probabilities, thereby producing more reliable and calibrated response scores in computing the uncertainty of generations. Our extensive experiments across multiple datasets show that LARS substantially outperforms existing scoring functions considering various probability-based UE methods.

Federated Learning (FL) allows clients to train a model collaboratively without sharing their private data. One key challenge in practical FL systems is data heterogeneity, particularly in handling clients with rare data, also referred to as Mavericks. These clients own one or more data classes exclusively, and the model performance becomes poor without their participation. Thus, utilizing Mavericks throughout training is crucial. In this paper, we first design a Maverick-aware Shapley valuation that fairly evaluates the contribution of Mavericks. The main idea is to compute the clients' Shapley values (SV) class-wise, i.e., per label. Next, we propose FedMS, a Maverick-Shapley client selection mechanism for FL that intelligently selects the clients that contribute the most in each round, by employing our Maverick-aware SV-based contribution score. We show that, compared to an extensive list of baselines, FedMS achieves better model performance and fairer Shapley Rewards distribution.

Diffusion-based generative models represent a forefront direction in generative AI research today. Recent studies in physics have suggested that the renormalization group (RG) can be conceptualized as a diffusion process. This insight motivates us to develop a novel diffusion-based generative model by reversing the momentum-space RG flow. We establish a framework that interprets RG flow as optimal transport gradient flow, which minimizes a functional analogous to the Kullback-Leibler divergence, thereby bridging statistical physics and information theory. Our model applies forward and reverse diffusion processes in Fourier space, exploiting the sparse representation of natural images in this domain to efficiently separate signal from noise and manage image features across scales. By introducing a scale-dependent noise schedule informed by a dispersion relation, the model optimizes denoising performance and image generation in Fourier space, taking advantage of the distinct separation of macro and microscale features. Experimental validations on standard datasets demonstrate the model's capability to generate high-quality images while significantly reducing training time compared to existing image-domain diffusion models. This approach not only enhances our understanding of the generative processes in images but also opens new pathways for research in generative AI, leveraging the convergence of theoretical physics, optimal transport, and machine learning principles., Comment: 18 pages, 7 figures

Generative Large Language Models (LLMs) are widely utilized for their excellence in various tasks. However, their tendency to produce inaccurate or misleading outputs poses a potential risk, particularly in high-stakes environments. Therefore, estimating the correctness of generative LLM outputs is an important task for enhanced reliability. Uncertainty Estimation (UE) in generative LLMs is an evolving domain, where SOTA probability-based methods commonly employ length-normalized scoring. In this work, we propose Meaning-Aware Response Scoring (MARS) as an alternative to length-normalized scoring for UE methods. MARS is a novel scoring function that considers the semantic contribution of each token in the generated sequence in the context of the question. We demonstrate that integrating MARS into UE methods results in a universal and significant improvement in UE performance. We conduct experiments using three distinct closed-book question-answering datasets across five popular pre-trained LLMs. Lastly, we validate the efficacy of MARS on a Medical QA dataset. Code can be found https://github.com/Ybakman/LLM_Uncertainity.

Federated Learning (FL) systems are vulnerable to adversarial attacks, where malicious clients submit poisoned models to prevent the global model from converging or plant backdoors to induce the global model to misclassify some samples. Current defense methods fall short in real-world FL systems, as they either rely on impractical prior knowledge or introduce accuracy loss even when no attack happens. Also, these methods do not offer a protocol for verifying the execution, leaving participants doubtful about the correct execution of the mechanism. To address these issues, we propose a novel anomaly detection strategy designed for real-world FL systems. Our approach activates the defense only upon occurrence of attacks, and removes malicious models accurately, without affecting the benign ones. Additionally, our approach incorporates zero-knowledge proofs to ensure the integrity of defense mechanisms. Experimental results demonstrate the effectiveness of our approach in enhancing the security of FL systems against adversarial attacks.

This paper introduces FedSecurity, an end-to-end benchmark that serves as a supplementary component of the FedML library for simulating adversarial attacks and corresponding defense mechanisms in Federated Learning (FL). FedSecurity eliminates the need for implementing the fundamental FL procedures, e.g., FL training and data loading, from scratch, thus enables users to focus on developing their own attack and defense strategies. It contains two key components, including FedAttacker that conducts a variety of attacks during FL training, and FedDefender that implements defensive mechanisms to counteract these attacks. FedSecurity has the following features: i) It offers extensive customization options to accommodate a broad range of machine learning models (e.g., Logistic Regression, ResNet, and GAN) and FL optimizers (e.g., FedAVG, FedOPT, and FedNOVA); ii) it enables exploring the effectiveness of attacks and defenses across different datasets and models; and iii) it supports flexible configuration and customization through a configuration file and some APIs. We further demonstrate FedSecurity's utility and adaptability through federated training of Large Language Models (LLMs) to showcase its potential on a wide range of complex applications.

We consider a project (model) owner that would like to train a model by utilizing the local private data and compute power of interested data owners, i.e., trainers. Our goal is to design a data marketplace for such decentralized collaborative/federated learning applications that simultaneously provides i) proof-of-contribution based reward allocation so that the trainers are compensated based on their contributions to the trained model; ii) privacy-preserving decentralized model training by avoiding any data movement from data owners; iii) robustness against malicious parties (e.g., trainers aiming to poison the model); iv) verifiability in the sense that the integrity, i.e., correctness, of all computations in the data market protocol including contribution assessment and outlier detection are verifiable through zero-knowledge proofs; and v) efficient and universal design. We propose a blockchain-based marketplace design to achieve all five objectives mentioned above. In our design, we utilize a distributed storage infrastructure and an aggregator aside from the project owner and the trainers. The aggregator is a processing node that performs certain computations, including assessing trainer contributions, removing outliers, and updating hyper-parameters. We execute the proposed data market through a blockchain smart contract. The deployed smart contract ensures that the project owner cannot evade payment, and honest trainers are rewarded based on their contributions at the end of training. Finally, we implement the building blocks of the proposed data market and demonstrate their applicability in practical scenarios through extensive experiments.

We consider a foundational unsupervised learning task of $k$-means data clustering, in a federated learning (FL) setting consisting of a central server and many distributed clients. We develop SecFC, which is a secure federated clustering algorithm that simultaneously achieves 1) universal performance: no performance loss compared with clustering over centralized data, regardless of data distribution across clients; 2) data privacy: each client's private data and the cluster centers are not leaked to other clients and the server. In SecFC, the clients perform Lagrange encoding on their local data and share the coded data in an information-theoretically private manner; then leveraging the algebraic structure of the coding, the FL network exactly executes the Lloyd's $k$-means heuristic over the coded data to obtain the final clustering. Experiment results on synthetic and real datasets demonstrate the universally superior performance of SecFC for different data distributions across clients, and its computational practicality for various combinations of system parameters. Finally, we propose an extension of SecFC to further provide membership privacy for all data points.

We consider a network consisting of a single source and $n$ receiver nodes that are grouped into equal-sized clusters. We use cluster heads in each cluster to facilitate communication between the source and the nodes within that cluster. Inside clusters, nodes are connected to each other according to a given network topology. Based on the connectivity among the nodes, each node relays its current stored version of the source update to its neighboring nodes by $local$ $gossiping$. We use the $version$ $age$ metric to assess information freshness at the nodes. We consider disconnected, ring, and fully connected network topologies for each cluster. For each network topology, we characterize the average version age at each node and find the average version age scaling as a function of the network size $n$. Our results indicate that per node average version age scalings of $O(\sqrt{n})$, $O(n^{\frac{1}{3}})$, and $O(\log n)$ are achievable in disconnected, ring, and fully connected cluster models, respectively. Next, we increase connectivity in the network and allow gossiping among the cluster heads to improve version age at the nodes. With that, we show that when the cluster heads form a ring network among themselves, we obtain per node average version age scalings of $O(n^{\frac{1}{3}})$, $O(n^{\frac{1}{4}})$, and $O(\log n)$ in disconnected, ring, and fully connected cluster models, respectively. Next, focusing on a ring network topology in each cluster, we introduce hierarchy to the considered clustered gossip network model and show that when we employ two levels of hierarchy, we can achieve the same $O(n^{\frac{1}{4}})$ scaling without using dedicated cluster heads. We generalize this result for $h$ levels of hierarchy and show that per user average version age scaling of $O(n^{\frac{1}{2h}})$ is achievable in the case of a ring network in each cluster across all hierarchy levels., Comment: arXiv admin note: substantial text overlap with arXiv:2105.02867

We consider the binary freshness metric for gossip networks that consist of a single source and $n$ end-nodes, where the end-nodes are allowed to share their stored versions of the source information with the other nodes. We develop recursive equations that characterize binary freshness in arbitrarily connected gossip networks using the stochastic hybrid systems (SHS) approach. Next, we study binary freshness in several structured gossip networks, namely disconnected, ring and fully connected networks. We show that for both disconnected and ring network topologies, when the number of nodes gets large, the binary freshness of a node decreases down to 0 as $n^{-1}$, but the freshness is strictly larger for the ring topology. We also show that for the fully connected topology, the rate of decrease to 0 is slower, and it takes the form of $n^{-\rho}$ for a $\rho$ smaller than 1, when the update rates of the source and the end-nodes are sufficiently large. Finally, we study the binary freshness metric for clustered gossip networks, where multiple clusters of structured gossip networks are connected to the source node through designated access nodes, i.e., cluster heads. We characterize the binary freshness in such networks and numerically study how the optimal cluster sizes change with respect to the update rates in the system.

We consider a network consisting of a single source and $n$ receiver nodes that are grouped into $m$ equal size communities, i.e., clusters, where each cluster includes $k$ nodes and is served by a dedicated cluster head. The source node keeps versions of an observed process and updates each cluster through the associated cluster head. Nodes within each cluster are connected to each other according to a given network topology. Based on this topology, each node relays its current update to its neighboring nodes by $local$ $gossiping$. We use the $version$ $age$ metric to quantify information timeliness at the receiver nodes. We consider disconnected, ring, and fully connected network topologies for each cluster. For each of these network topologies, we characterize the average version age at each node and find the version age scaling as a function of the network size $n$. Our results indicate that per node version age scalings of $O(\sqrt{n})$, $O(n^{\frac{1}{3}})$, and $O(\log n)$ are achievable in disconnected, ring, and fully connected cluster models, respectively. Finally, through numerical evaluations, we determine the version age-optimum $(m,k)$ pairs as a function of the source, cluster head, and node update rates.

Distributed implementations are crucial in speeding up large scale machine learning applications. Distributed gradient descent (GD) is widely employed to parallelize the learning task by distributing the dataset across multiple workers. A significant performance bottleneck for the per-iteration completion time in distributed synchronous GD is $straggling$ workers. Coded distributed computation techniques have been introduced recently to mitigate stragglers and to speed up GD iterations by assigning redundant computations to workers. In this paper, we consider gradient coding (GC), and propose a novel dynamic GC scheme, which assigns redundant data to workers to acquire the flexibility to dynamically choose from among a set of possible codes depending on the past straggling behavior. In particular, we consider GC with clustering, and regulate the number of stragglers in each cluster by dynamically forming the clusters at each iteration; hence, the proposed scheme is called $GC$ $with$ $dynamic$ $clustering$ (GC-DC). Under a time-correlated straggling behavior, GC-DC gains from adapting to the straggling behavior over time such that, at each iteration, GC-DC aims at distributing the stragglers across clusters as uniformly as possible based on the past straggler behavior. For both homogeneous and heterogeneous worker models, we numerically show that GC-DC provides significant improvements in the average per-iteration completion time without an increase in the communication load compared to the original GC scheme., Comment: arXiv admin note: substantial text overlap with arXiv:2011.01922

We consider a federated learning framework in which a parameter server (PS) trains a global model by using $n$ clients without actually storing the client data centrally at a cloud server. Focusing on a setting where the client datasets are fast changing and highly temporal in nature, we investigate the timeliness of model updates and propose a novel timely communication scheme. Under the proposed scheme, at each iteration, the PS waits for $m$ available clients and sends them the current model. Then, the PS uses the local updates of the earliest $k$ out of $m$ clients to update the global model at each iteration. We find the average age of information experienced by each client and numerically characterize the age-optimal $m$ and $k$ values for a given $n$. Our results indicate that, in addition to ensuring timeliness, the proposed communication scheme results in significantly smaller average iteration times compared to random client selection without hurting the convergence of the global learning task.

In distributed synchronous gradient descent (GD) the main performance bottleneck for the per-iteration completion time is the slowest \textit{straggling} workers. To speed up GD iterations in the presence of stragglers, coded distributed computation techniques are implemented by assigning redundant computations to workers. In this paper, we propose a novel gradient coding (GC) scheme that utilizes dynamic clustering, denoted by GC-DC, to speed up the gradient calculation. Under time-correlated straggling behavior, GC-DC aims at regulating the number of straggling workers in each cluster based on the straggler behavior in the previous iteration. We numerically show that GC-DC provides significant improvements in the average completion time (of each iteration) with no increase in the communication load compared to the original GC scheme.

Coded computation can be used to speed up distributed learning in the presence of straggling workers. Partial recovery of the gradient vector can further reduce the computation time at each iteration; however, this can result in biased estimators, which may slow down convergence, or even cause divergence. Estimator bias will be particularly prevalent when the straggling behavior is correlated over time, which results in the gradient estimators being dominated by a few fast servers. To mitigate biased estimators, we design a $timely$ dynamic encoding framework for partial recovery that includes an ordering operator that changes the codewords and computation orders at workers over time. To regulate the recovery frequencies, we adopt an $age$ metric in the design of the dynamic encoding scheme. We show through numerical results that the proposed dynamic encoding strategy increases the timeliness of the recovered computations, which as a result, reduces the bias in model updates, and accelerates the convergence compared to the conventional static partial recovery schemes.

An information source generates independent and identically distributed status update messages from an observed random phenomenon which takes $n$ distinct values based on a given pmf. These update packets are encoded at the transmitter node to be sent to a receiver node which wants to track the observed random variable with as little age as possible. The transmitter node implements a selective $k$ encoding policy such that rather than encoding all possible $n$ realizations, the transmitter node encodes the most probable $k$ realizations. We consider three different policies regarding the remaining $n-k$ less probable realizations: $highest$ $k$ $selective$ $encoding$ which disregards whenever a realization from the remaining $n-k$ values occurs; $randomized$ $selective$ $encoding$ which encodes and sends the remaining $n-k$ realizations with a certain probability to further inform the receiver node at the expense of longer codewords for the selected $k$ realizations; and $highest$ $k$ $selective$ $encoding$ $with$ $an$ $empty$ $symbol$ which sends a designated empty symbol when one of the remaining $n-k$ realizations occurs. For all of these three encoding schemes, we find the average age and determine the age-optimal real codeword lengths, including the codeword length for the empty symbol in the case of the latter scheme, such that the average age at the receiver node is minimized. Through numerical evaluations for arbitrary pmfs, we show that these selective encoding policies result in a lower average age than encoding every realization, and find the corresponding age-optimal $k$ values., Comment: Submitted for publication, April 2020. Some text overlap with its conference version arXiv:2001.09975

We study age of information in a status updating system that consists of a single sampler, i.e., source node, that sends time-sensitive status updates to a single monitor node through a server node. We first consider a Gilbert-Elliot service profile at the server node. In this model, service times at the server node follow a finite state Markov chain with two states: ${bad}$ state $b$ and ${good}$ state $g$ where the server is faster in state $g$. We determine the time average age experienced by the monitor node and characterize the age-optimal state transition matrix $P$ with and without an average cost constraint on the service operation. Next, we consider a Gilbert-Elliot sampling profile at the source. In this model, the interarrival times follow a finite state Markov chain with two states: ${bad}$ state $b$ and ${good}$ state $g$ where samples are more frequent in state $g$. We find the time average age experienced by the monitor node and characterize the age-optimal state transition matrix $P$.

We consider a system in which an information source generates independent and identically distributed status update packets from an observed phenomenon that takes $n$ possible values based on a given pmf. These update packets are encoded at the transmitter node to be sent to the receiver node. Instead of encoding all $n$ possible realizations, the transmitter node only encodes the most probable $k$ realizations and disregards whenever a realization from the remaining $n-k$ values occurs. We find the average age and determine the age-optimal real codeword lengths such that the average age at the receiver node is minimized. Through numerical evaluations for arbitrary pmfs, we show that this selective encoding policy results in a lower average age than encoding every realization and find the age-optimal $k$. We also analyze a randomized selective encoding policy in which the remaining $n-k$ realizations are encoded and sent with a certain probability to further inform the receiver at the expense of longer codewords for the selected $k$ realizations.

We consider a status update system in which the update packets need to be processed to extract the embedded useful information. The source node sends the acquired information to a computation unit (CU) which consists of a master node and $n$ worker nodes. The master node distributes the received computation task to the worker nodes. Upon computation, the master node aggregates the results and sends them back to the source node to keep it \emph{updated}. We investigate the age performance of uncoded and coded (repetition coded, MDS coded, and multi-message MDS (MM-MDS) coded) schemes in the presence of stragglers under i.i.d.~exponential transmission delays and i.i.d~shifted exponential computation times. We show that asymptotically MM-MDS coded scheme outperforms the other schemes. Furthermore, we characterize the optimal codes such that the average age is minimized., Comment: Submitted for publication

Given $n$ randomly located source-destination (S-D) pairs on a fixed area network that want to communicate with each other, we study the age of information with a particular focus on its scaling as the network size $n$ grows. We propose a three-phase transmission scheme that utilizes \textit{hierarchical cooperation} between users along with \textit{mega update packets} and show that an average age scaling of $O(n^{\alpha(h)}\log n)$ per-user is achievable where $h$ denotes the number of hierarchy levels and $\alpha(h) = \frac{1}{3\cdot2^h+1}$ which tends to $0$ as $h$ increases such that asymptotically average age scaling of the proposed scheme is $O(\log n)$. To the best of our knowledge, this is the best average age scaling result in a status update system with multiple S-D pairs.

We consider the age of information in a multicast network where there is a single source node that sends time-sensitive updates to $n$ receiver nodes. Each status update is one of two kinds: type I or type II. To study the age of information experienced by the receiver nodes for both types of updates, we consider two cases: update streams are generated by the source node at-will and update streams arrive exogenously to the source node. We show that using an earliest $k_1$ and $k_2$ transmission scheme for type I and type II updates, respectively, the age of information of both update streams at the receiver nodes can be made a constant independent of $n$. In particular, the source node transmits each type I update packet to the earliest $k_1$ and each type II update packet to the earliest $k_2$ of $n$ receiver nodes. We determine the optimum $k_1$ and $k_2$ stopping thresholds for arbitrary shifted exponential link delays to individually and jointly minimize the average age of both update streams and characterize the pareto optimal curve for the two ages.

We consider the age of information in a multihop multicast network where there is a single source node sending time-sensitive updates to $n^L$ end nodes, and $L$ denotes the number of hops. In the first hop, the source node sends updates to $n$ first-hop receiver nodes, and in the second hop each first-hop receiver node relays the update packets that it has received to $n$ further users that are connected to it. This network architecture continues in further hops such that each receiver node in hop $\ell$ is connected to $n$ further receiver nodes in hop $\ell+1$. We study the age of information experienced by the end nodes, and in particular, its scaling as a function of $n$. We show that, using an earliest $k$ transmission scheme in each hop, the age of information at the end nodes can be made a constant independent of $n$. In particular, the source node transmits each update packet to the earliest $k_1$ of the $n$ first-hop nodes, and each first-hop node that receives the update relays it to the earliest $k_2$ out of $n$ second-hop nodes that are connected to it and so on. We determine the optimum $k_\ell$ stopping value for each hop $\ell$ for arbitrary shifted exponential link delays., Comment: Submitted for publication, December 2018

We study age of information in a multiple source-multiple destination setting with a focus on its scaling in large wireless networks. There are $n$ nodes that are randomly paired with each other on a fixed area to form $n$ source-destination (S-D) pairs. We propose a three-phase transmission scheme which utilizes local cooperation between the nodes by forming what we call mega update packets to serve multiple S-D pairs at once. We show that under the proposed scheme average age of an S-D pair scales as $O(n^{\frac{1}{4}})$ as the number of users, $n$, in the network grows. To the best of our knowledge, this is the best age scaling result for a multiple source-multiple destination setting.