Search Results for author: Theodoros Rekatsinas

Found 23 papers, 7 papers with code

Co-design Hardware and Algorithm for Vector Search

1 code implementation19 Jun 2023 Wenqi Jiang, Shigang Li, Yu Zhu, Johannes De Fine Licht, Zhenhao He, Runbin Shi, Cedric Renggli, Shuai Zhang, Theodoros Rekatsinas, Torsten Hoefler, Gustavo Alonso

Vector search has emerged as the foundation for large-scale information retrieval and machine learning systems, with search engines like Google and Bing processing tens of thousands of queries per second on petabyte-scale document datasets by evaluating vector similarities between encoded query texts and web documents.

Information Retrieval Retrieval

Repeated Random Sampling for Minimizing the Time-to-Accuracy of Learning

no code implementations28 May 2023 Patrik Okanovic, Roger Waleffe, Vasilis Mageirakos, Konstantinos E. Nikolakakis, Amin Karbasi, Dionysis Kalogerias, Nezihe Merve Gürel, Theodoros Rekatsinas

Methods for carefully selecting or generating a small set of training data to learn from, i. e., data pruning, coreset selection, and data distillation, have been shown to be effective in reducing the ever-increasing cost of training neural networks.

Data Compression

Growing and Serving Large Open-domain Knowledge Graphs

no code implementations16 May 2023 Ihab F. Ilyas, JP Lacerda, Yunyao Li, Umar Farooq Minhas, Ali Mousavi, Jeffrey Pound, Theodoros Rekatsinas, Chiraag Sumanth

We then describe how our platform, including graph embeddings, can be leveraged to create a Semantic Annotation service that links unstructured Web documents to entities in our KG.

Entity Linking Fact Verification +2

High-Throughput Vector Similarity Search in Knowledge Graphs

no code implementations4 Apr 2023 Jason Mohoney, Anil Pacaci, Shihabur Rahman Chowdhury, Ali Mousavi, Ihab F. Ilyas, Umar Farooq Minhas, Jeffrey Pound, Theodoros Rekatsinas

Motivated by the tasks of finding related KG queries and entities for past KG query workloads, we focus on hybrid vector similarity search (hybrid queries for short) where part of the query corresponds to vector similarity search and part of the query corresponds to predicates over relational attributes associated with the underlying data vectors.

Knowledge Graphs Vocal Bursts Intensity Prediction

Saga: A Platform for Continuous Construction and Serving of Knowledge At Scale

no code implementations15 Apr 2022 Ihab F. Ilyas, Theodoros Rekatsinas, Vishnu Konda, Jeffrey Pound, Xiaoguang Qi, Mohamed Soliman

We introduce Saga, a next-generation knowledge construction and serving platform for powering knowledge-based applications at industrial scale.

graph construction

Ember: No-Code Context Enrichment via Similarity-Based Keyless Joins

1 code implementation2 Jun 2021 Sahaana Suri, Ihab F. Ilyas, Christopher Ré, Theodoros Rekatsinas

Context enrichment, or rebuilding fragmented context, using keyless joins is an implicit or explicit step in machine learning (ML) pipelines over structured data sources.

Question Answering Representation Learning

Marius: Learning Massive Graph Embeddings on a Single Machine

1 code implementation20 Jan 2021 Jason Mohoney, Roger Waleffe, Yiheng Xu, Theodoros Rekatsinas, Shivaram Venkataraman

We propose a new framework for computing the embeddings of large-scale graphs on a single machine.

Graph Embedding

Principal Component Networks: Parameter Reduction Early in Training

no code implementations23 Jun 2020 Roger Waleffe, Theodoros Rekatsinas

Recent works show that overparameterized networks contain small subnetworks that exhibit comparable accuracy to the full model when trained in isolation.

Record fusion: A learning approach

no code implementations18 Jun 2020 Alireza Heidari, George Michalopoulos, Shrinu Kushagra, Ihab F. Ilyas, Theodoros Rekatsinas

We use this feature vector alongwith the ground-truth information to learn a classifier for each of the attributes of the database.

Attribute

An Empirical Analysis of the Impact of Data Augmentation on Knowledge Distillation

no code implementations6 Jun 2020 Deepan Das, Haley Massa, Abhimanyu Kulkarni, Theodoros Rekatsinas

Generalization Performance of Deep Learning models trained using Empirical Risk Minimization can be improved significantly by using Data Augmentation strategies such as simple transformations, or using Mixed Samples.

Data Augmentation Knowledge Distillation

On Robust Mean Estimation under Coordinate-level Corruption

no code implementations10 Feb 2020 Zifan Liu, Jongho Park, Theodoros Rekatsinas, Christos Tzamos

We study the problem of robust mean estimation and introduce a novel Hamming distance-based measure of distribution shift for coordinate-level corruptions.

Matrix Completion

Fine-Grained Object Detection over Scientific Document Images with Region Embeddings

no code implementations28 Oct 2019 Ankur Goswami, Joshua McGrath, Shanan Peters, Theodoros Rekatsinas

We also present a region embedding model that uses the convolutional maps of a proposal's neighbors as context to produce an embedding for each proposal.

Object object-detection +1

Approximate Inference in Structured Instances with Noisy Categorical Observations

no code implementations29 Jun 2019 Alireza Heidari, Ihab F. Ilyas, Theodoros Rekatsinas

We study the problem of recovering the latent ground truth labeling of a structured instance with categorical random variables in the presence of noisy observations.

Clustering Structured Prediction

Data-Dependent Differentially Private Parameter Learning for Directed Graphical Models

no code implementations ICML 2020 Amrita Roy Chowdhury, Theodoros Rekatsinas, Somesh Jha

Our solution optimizes for the utility of inference queries over the DGM and \textit{adds noise that is customized to the properties of the private input dataset and the graph structure of the DGM}.

Learning Functional Dependencies with Sparse Regression

no code implementations4 May 2019 Zhihan Guo, Theodoros Rekatsinas

We show that discovering FDs from a noisy dataset is equivalent to learning the structure of a graphical model over binary random variables, where each random variable corresponds to a functional of the dataset attributes.

regression

Unsupervised Functional Dependency Discovery for Data Preparation

no code implementations ICLR Workshop LLD 2019 Zhihan Guo, Theodoros Rekatsinas

We study the problem of functional dependency (FD) discovery to impose domain knowledge for downstream data preparation tasks.

regression

Multi-relational Learning Using Weighted Tensor Decomposition with Modular Loss

no code implementations7 Mar 2013 Ben London, Theodoros Rekatsinas, Bert Huang, Lise Getoor

For the typical cases of real-valued functions and binary relations, we propose several loss functions and derive the associated parameter gradients.

Relational Reasoning Tensor Decomposition

Cannot find the paper you are looking for? You can Submit a new open access paper.