The more AI-assisted decisions affect people's lives, the more important the fairness of such decisions becomes. In this chapter, we provide an introduction to research on fairness in machine learning. We explain the main fairness definitions and strategies for achieving fairness using concrete examples and place fairness research in the European context. Our contribution is aimed at an interdisciplinary audience and therefore avoids mathematical formulation but emphasizes visualizations and examples. -- Je mehr KI-gest\"utzte Entscheidungen das Leben von Menschen betreffen, desto wichtiger ist die Fairness solcher Entscheidungen. In diesem Kapitel geben wir eine Einf\"uhrung in die Forschung zu Fairness im maschinellen Lernen. Wir erkl\"aren die wesentlichen Fairness-Definitionen und Strategien zur Erreichung von Fairness anhand konkreter Beispiele und ordnen die Fairness-Forschung in den europ\"aischen Kontext ein. Unser Beitrag richtet sich dabei an ein interdisziplin\"ares Publikum und verzichtet daher auf die mathematische Formulierung sondern betont Visualisierungen und Beispiele.
相關內容
The Tor Network has been a significant part of the Internet for years. Tor was originally started in the Naval Research Laboratory for anonymous Internet browsing and Internet-based communication. From being used for anonymous communications, it has now segmented into various other use-cases like censorship circumvention, performing illegal activities, etc. In this paper, we perform empirical measurements on the Tor network to analyze the trends in Tor over the years. We gather our measurements data through our measurement scripts, past research in this domain, and aggregated data provided by the Tor metrics directory. We use this data to analyze trends and understand the incidents that caused fluctuations in the trends of different data parameters. We collect measurements data for Tor parameters like Tor users, onion services, Tor relays, and bridges, etc. We also study censorshiprelated events and study trends by analyzing censorship-related metrics. Finally, we touch upon the location diversity in Tor and study how the Tor circuit selection and construction are impacted by the bandwidth distribution of Tor relays across geographies.
In recent years, Large Language Models (LLMs) have demonstrated remarkable potential across various downstream tasks. LLM-integrated frameworks, which serve as the essential infrastructure, have given rise to many LLM-integrated web apps. However, some of these frameworks suffer from Remote Code Execution (RCE) vulnerabilities, allowing attackers to execute arbitrary code on apps' servers remotely via prompt injections. Despite the severity of these vulnerabilities, no existing work has been conducted for a systematic investigation of them. This leaves a great challenge on how to detect vulnerabilities in frameworks as well as LLM-integrated apps in real-world scenarios. To fill this gap, we present two novel strategies, including 1) a static analysis-based tool called LLMSmith to scan the source code of the framework to detect potential RCE vulnerabilities and 2) a prompt-based automated testing approach to verify the vulnerability in LLM-integrated web apps. We discovered 13 vulnerabilities in 6 frameworks, including 12 RCE vulnerabilities and 1 arbitrary file read/write vulnerability. 11 of them are confirmed by the framework developers, resulting in the assignment of 7 CVE IDs. After testing 51 apps, we found vulnerabilities in 17 apps, 16 of which are vulnerable to RCE and 1 to SQL injection. We responsibly reported all 17 issues to the corresponding developers and received acknowledgments. Furthermore, we amplify the attack impact beyond achieving RCE by allowing attackers to exploit other app users (e.g. app responses hijacking, user API key leakage) without direct interaction between the attacker and the victim. Lastly, we propose some mitigating strategies for improving the security awareness of both framework and app developers, helping them to mitigate these risks effectively.
This paper addresses the problem of Age-of-Information (AoI) in UAV-assisted networks. Our objective is to minimize the expected AoI across devices by optimizing UAVs' stopping locations and device selection probabilities. To tackle this problem, we first derive a closed-form expression of the expected AoI that involves the probabilities of selection of devices. Then, we formulate the problem as a non-convex minimization subject to quality of service constraints. Since the problem is challenging to solve, we propose an Ensemble Deep Neural Network (EDNN) based approach which takes advantage of the dual formulation of the studied problem. Specifically, the Deep Neural Networks (DNNs) in the ensemble are trained in an unsupervised manner using the Lagrangian function of the studied problem. Our experiments show that the proposed EDNN method outperforms traditional DNNs in reducing the expected AoI, achieving a remarkable reduction of $29.5\%$.
Recently, Kirkpatrick et al. [ALT 2019] and Fallat et al. [JMLR 2023] introduced non-clashing teaching and showed it to be the most efficient machine teaching model satisfying the benchmark for collusion-avoidance set by Goldman and Mathias. A teaching map $T$ for a concept class $\cal{C}$ assigns a (teaching) set $T(C)$ of examples to each concept $C \in \cal{C}$. A teaching map is non-clashing if no pair of concepts are consistent with the union of their teaching sets. The size of a non-clashing teaching map (NCTM) $T$ is the maximum size of a $T(C)$, $C \in \cal{C}$. The non-clashing teaching dimension NCTD$(\cal{C})$ of $\cal{C}$ is the minimum size of an NCTM for $\cal{C}$. NCTM$^+$ and NCTD$^+(\cal{C})$ are defined analogously, except the teacher may only use positive examples. We study NCTMs and NCTM$^+$s for the concept class $\mathcal{B}(G)$ consisting of all balls of a graph $G$. We show that the associated decision problem {\sc B-NCTD$^+$} for NCTD$^+$ is NP-complete in split, co-bipartite, and bipartite graphs. Surprisingly, we even prove that, unless the ETH fails, {\sc B-NCTD$^+$} does not admit an algorithm running in time $2^{2^{o(vc)}}\cdot n^{O(1)}$, nor a kernelization algorithm outputting a kernel with $2^{o(vc)}$ vertices, where vc is the vertex cover number of $G$. These are extremely rare results: it is only the second (fourth, resp.) problem in NP to admit a double-exponential lower bound parameterized by vc (treewidth, resp.), and only one of very few problems to admit an ETH-based conditional lower bound on the number of vertices in a kernel. We complement these lower bounds with matching upper bounds. For trees, interval graphs, cycles, and trees of cycles, we derive NCTM$^+$s or NCTMs for $\mathcal{B}(G)$ of size proportional to its VC-dimension. For Gromov-hyperbolic graphs, we design an approximate NCTM$^+$ for $\mathcal{B}(G)$ of size 2.
We show that participating in federated learning can be detrimental to group fairness. In fact, the bias of a few parties against under-represented groups (identified by sensitive attributes such as gender or race) can propagate through the network to all the parties in the network. We analyze and explain bias propagation in federated learning on naturally partitioned real-world datasets. Our analysis reveals that biased parties unintentionally yet stealthily encode their bias in a small number of model parameters, and throughout the training, they steadily increase the dependence of the global model on sensitive attributes. What is important to highlight is that the experienced bias in federated learning is higher than what parties would otherwise encounter in centralized training with a model trained on the union of all their data. This indicates that the bias is due to the algorithm. Our work calls for auditing group fairness in federated learning and designing learning algorithms that are robust to bias propagation.
Voting mechanisms play a crucial role in decentralized governance of blockchain systems. Liquid democracy, also known as delegative voting, allows voters to vote directly or delegate their voting power to others, thereby contributing to the resolution of problems such as low voter turnout. In recent years, liquid democracy has been widely adopted by Delegated-Proof-of-Stake (DPoS) blockchains and implemented successfully on platforms with millions of users. However, little is known regarding the characteristics and actual effectiveness of liquid democracy in decentralized governance. This paper explored for the first time the practical implementation of liquid democracy in DPoS blockchain systems. Using actual data collected from two major DPoS blockchains, EOS and Steem, our study compared and evaluated the participation of different types of users of DPoS blockchain systems in liquid democracy, as well as extracting and analyzing the delegation chains and networks formed during the process of liquid democracy within the systems. We believe that the findings of this paper will contribute to further studies on the design and implementation of liquid democracy and other voting mechanisms in decentralized governance.
We describe a class of tasks called decision-oriented dialogues, in which AI assistants must collaborate with one or more humans via natural language to help them make complex decisions. We formalize three domains in which users face everyday decisions: (1) choosing an assignment of reviewers to conference papers, (2) planning a multi-step itinerary in a city, and (3) negotiating travel plans for a group of friends. In each of these settings, AI assistants and users have disparate abilities that they must combine to arrive at the best decision: assistants can access and process large amounts of information, while users have preferences and constraints external to the system. For each task, we build a dialogue environment where agents receive a reward based on the quality of the final decision they reach. Using these environments, we collect human-human dialogues with humans playing the role of assistant. To compare how current AI assistants communicate in these settings, we present baselines using large language models in self-play. Finally, we highlight a number of challenges models face in decision-oriented dialogues, ranging from efficient communication to reasoning and optimization, and release our environments as a testbed for future modeling work.
The new era of technology has brought us to the point where it is convenient for people to share their opinions over an abundance of platforms. These platforms have a provision for the users to express themselves in multiple forms of representations, including text, images, videos, and audio. This, however, makes it difficult for users to obtain all the key information about a topic, making the task of automatic multi-modal summarization (MMS) essential. In this paper, we present a comprehensive survey of the existing research in the area of MMS.
Knowledge graphs are important resources for many artificial intelligence tasks but often suffer from incompleteness. In this work, we propose to use pre-trained language models for knowledge graph completion. We treat triples in knowledge graphs as textual sequences and propose a novel framework named Knowledge Graph Bidirectional Encoder Representations from Transformer (KG-BERT) to model these triples. Our method takes entity and relation descriptions of a triple as input and computes scoring function of the triple with the KG-BERT language model. Experimental results on multiple benchmark knowledge graphs show that our method can achieve state-of-the-art performance in triple classification, link prediction and relation prediction tasks.
We investigate the problem of automatically determining what type of shoe left an impression found at a crime scene. This recognition problem is made difficult by the variability in types of crime scene evidence (ranging from traces of dust or oil on hard surfaces to impressions made in soil) and the lack of comprehensive databases of shoe outsole tread patterns. We find that mid-level features extracted by pre-trained convolutional neural nets are surprisingly effective descriptors for this specialized domains. However, the choice of similarity measure for matching exemplars to a query image is essential to good performance. For matching multi-channel deep features, we propose the use of multi-channel normalized cross-correlation and analyze its effectiveness. Our proposed metric significantly improves performance in matching crime scene shoeprints to laboratory test impressions. We also show its effectiveness in other cross-domain image retrieval problems: matching facade images to segmentation labels and aerial photos to map images. Finally, we introduce a discriminatively trained variant and fine-tune our system through our proposed metric, obtaining state-of-the-art performance.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191