Given a database with missing or uncertain content, our goal is to correct and fill the database by extracting specific information from a large corpus such as the Web, and to do so under resource limitations. We formulate the information gathering task as a series of choices among alternative, resource-consuming actions and use reinforcement learning to select the best action at each time step. We use temporal difference q-learning method to train the function that selects these actions, and compare it to an online, errordriven algorithm called SampleRank. We present a system that finds information such as email, job title and department affiliation for the faculty at our university, and show that the learning-based approach accomplishes this task efficiently under a limited action budget. Our evaluations show that we can obtain 92.4% of the final F1, by only using 14.3% of all possible actions.

Federated Learning (FL) is quickly becoming a goto distributed training paradigm for users to jointly train a global model without physically sharing their data. Users can indirectly contribute to, and directly benefit from a much larger aggregate data corpus used to train the global model. However, literature on successful application of FL in real-world problem settings is somewhat sparse. In this paper, we describe our experience applying a FL based solution to the Named Entity Recognition (NER) task for an adverse event detection application in the context of mass scale vaccination programs. We present a comprehensive empirical analysis of various dimensions of benefits gained with FL based training. Furthermore, we investigate effects of tighter Differential Privacy (DP) constraints in highly sensitive settings where federation users must enforce Local DP to ensure strict privacy guarantees. We show that local DP can severely cripple the global model's prediction accuracy, thus disincentivizing users from participating in the federation. In response, we demonstrate how recent innovation on personalization methods can help significantly recover the lost accuracy. We focus our analysis on the Federated Fine-Tuning algorithm, FedFT, and prove that it is not PAC Identifiable, thus making it even more attractive for FL-based training.

Abstract. We present a general framework for the task of extracting specific information "on demand" from a large corpus such as the Web under resource-constraints. Given a database with missing or uncertain information, the proposed system automatically formulates queries, issues them to a search interface, selects a subset of the documents, extracts the required information from them, and fills the missing values in the original database. We also exploit inherent dependency within the data to obtain useful information with fewer computational resources. We build such a system in the citation database domain that extracts the missing publication years using limited resources from the Web. We discuss a probabilistic approach for this task and present first results. The main contribution of this paper is to propose a general, comprehensive architecture for designing a system adaptable to different domains.

No abstract

We present a new class of problems, called resource-bounded information gathering for correlation clustering. Our goal is to perform correlation clustering under circumstances in which accuracy may be improved by augmenting the given graph with additional information. This information is obtained by querying an external source under resource constraints. The problem is to develop the most effective query selection strategy to minimize some loss function on the resulting partitioning. We motivate the problem using an entity resolution task.