Search Results for author: Zohar Karnin

Found 31 papers, 4 papers with code

TADPOLE: Task ADapted Pre-Training via AnOmaLy DEtection

no code implementations EMNLP 2021 Vivek Madan, Ashish Khetan, Zohar Karnin

In this paper, we address the problem for the case when the downstream corpus is too small for additional pre-training.

Anomaly Detection Data Augmentation +1

Uncovering Drift in Textual Data: An Unsupervised Method for Detecting and Mitigating Drift in Machine Learning Models

no code implementations7 Sep 2023 Saeed Khaki, Akhouri Abhinav Aditya, Zohar Karnin, Lan Ma, Olivia Pan, Samarth Marudheri Chandrashekar

Our first step involves encoding a sample of production data as the target distribution, and the model training data as the reference distribution.

Representation Projection Invariance Mitigates Representation Collapse

no code implementations23 May 2022 Anastasia Razdaibiedina, Ashish Khetan, Zohar Karnin, Daniel Khashabi, Vishaal Kapoor, Vivek Madan

In this paper, we propose Representation Projection Invariance (REPINA), a novel regularization method to maintain the information content of representation and reduce representation collapse during fine-tuning by discouraging undesirable changes in the representations.

Pyramid-BERT: Reducing Complexity via Successive Core-set based Token Selection

no code implementations ACL 2022 Xin Huang, Ashish Khetan, Rene Bidart, Zohar Karnin

Transformer-based language models such as BERT have achieved the state-of-the-art performance on various NLP tasks, but are computationally prohibitive.

Amazon SageMaker Model Monitor: A System for Real-Time Insights into Deployed Machine Learning Models

no code implementations26 Nov 2021 David Nigenda, Zohar Karnin, Muhammad Bilal Zafar, Raghu Ramesha, Alan Tan, Michele Donini, Krishnaram Kenthapadi

With the increasing adoption of machine learning (ML) models and systems in high-stakes settings across different industries, guaranteeing a model's performance after deployment has become crucial.

BIG-bench Machine Learning

Improving Early Sepsis Prediction with Multi Modal Learning

no code implementations23 Jul 2021 Fred Qin, Vivek Madan, Ujjwal Ratan, Zohar Karnin, Vishaal Kapoor, Parminder Bhatia, Taha Kass-Hout

Clinical text provides essential information to estimate the severity of the sepsis in addition to structured clinical data.

Domain Adaptation via Anaomaly Detection

no code implementations1 Jan 2021 Vivek Madan, Ashish Khetan, Zohar Karnin

The need for such a method is clear as it is infeasible to collect a large pre-training corpus for every possible domain.

Anomaly Detection Domain Adaptation +1

GANMEX: Class-Targeted One-vs-One Attributions using GAN-based Model Explainability

no code implementations1 Jan 2021 Sheng-Min Shih, Pin-Ju Tien, Zohar Karnin

Our approach effectively selects the baseline as the closest realistic sample belong to the target class, which allows attribution methods to provide true one-vs-one explanations.

TabTransformer: Tabular Data Modeling Using Contextual Embeddings

11 code implementations11 Dec 2020 Xin Huang, Ashish Khetan, Milan Cvitkovic, Zohar Karnin

We propose TabTransformer, a novel deep tabular data modeling architecture for supervised and semi-supervised learning.

tabular-classification Unsupervised Pre-training

GANMEX: One-vs-One Attributions Guided by GAN-based Counterfactual Explanation Baselines

1 code implementation11 Nov 2020 Sheng-Min Shih, Pin-Ju Tien, Zohar Karnin

Attribution methods have been shown as promising approaches for identifying key features that led to learned model predictions.

counterfactual Counterfactual Explanation

Practical and sample efficient zero-shot HPO

no code implementations27 Jul 2020 Fela Winkelmolen, Nikita Ivkin, H. Furkan Bozkurt, Zohar Karnin

Zero-shot hyperparameter optimization (HPO) is a simple yet effective use of transfer learning for constructing a small list of hyperparameter (HP) configurations that complement each other.

Hyperparameter Optimization Transfer Learning

PruneNet: Channel Pruning via Global Importance

no code implementations22 May 2020 Ashish Khetan, Zohar Karnin

The methods that start with a pretrained network either prune channels uniformly across the layers or prune channels based on the basic statistics of the network parameters.

schuBERT: Optimizing Elements of BERT

no code implementations ACL 2020 Ashish Khetan, Zohar Karnin

In this work we revisit the architecture choices of BERT in efforts to obtain a lighter model.

Differentiable Architecture Compression

no code implementations ICLR 2020 Shashank Singh, Ashish Khetan, Zohar Karnin

In many learning situations, resources at inference time are significantly more constrained than resources at training time.

Image Classification Model Compression

Streaming Quantiles Algorithms with Small Space and Update Time

1 code implementation29 Jun 2019 Nikita Ivkin, Edo Liberty, Kevin Lang, Zohar Karnin, Vladimir Braverman

Approximating quantiles and distributions over streaming data has been studied for roughly two decades now.

Asymmetric Random Projections

no code implementations22 Jun 2019 Nick Ryder, Zohar Karnin, Edo Liberty

In many applications the data set to be projected is given to us in advance, yet the current RP techniques do not make use of information about the data.

General Classification regression

Discrepancy, Coresets, and Sketches in Machine Learning

no code implementations11 Jun 2019 Zohar Karnin, Edo Liberty

We provide general techniques for bounding the class discrepancy of machine learning problems.

BIG-bench Machine Learning Density Estimation

DARC: Differentiable ARchitecture Compression

no code implementations20 May 2019 Shashank Singh, Ashish Khetan, Zohar Karnin

In many learning situations, resources at inference time are significantly more constrained than resources at training time.

Image Classification Model Compression +1

Adaptive Feature Selection: Computationally Efficient Online Sparse Linear Regression under RIP

no code implementations ICML 2017 Satyen Kale, Zohar Karnin, Tengyuan Liang, Dávid Pál

Online sparse linear regression is an online problem where an algorithm repeatedly chooses a subset of coordinates to observe in an adversarially chosen feature vector, makes a real-valued prediction, receives the true label, and incurs the squared loss.

feature selection regression

One-Shot Session Recommendation Systems with Combinatorial Items

no code implementations5 Jul 2016 Yahel David, Dotan Di Castro, Zohar Karnin

Our optimization problem is formulated as an MDP where the action space is of a combinatorial nature as we recommend in each round, multiple items.

Recommendation Systems

How Many Folders Do You Really Need?

no code implementations29 Jun 2016 Mihajlo Grbovic, Guy Halawi, Zohar Karnin, Yoelle Maarek

We report how we have discovered that a set of 6 "latent" categories (one for human- and the others for machine-generated messages) can explain a significant portion of email traffic.

General Classification

Optimal Quantile Approximation in Streams

2 code implementations17 Mar 2016 Zohar Karnin, Kevin Lang, Edo Liberty

One of our contributions is a novel representation and modification of the widely used merge-and-reduce construction.

Data Structures and Algorithms

Copeland Dueling Bandits

no code implementations NeurIPS 2015 Masrour Zoghi, Zohar Karnin, Shimon Whiteson, Maarten de Rijke

A version of the dueling bandit problem is addressed in which a Condorcet winner may not exist.

Budget-Constrained Item Cold-Start Handling in Collaborative Filtering Recommenders via Optimal Design

no code implementations10 Jun 2014 Oren Anava, Shahar Golan, Nadav Golbandi, Zohar Karnin, Ronny Lempel, Oleg Rokhlenko, Oren Somekh

It is well known that collaborative filtering (CF) based recommender systems provide better modeling of users and items associated with considerable rating history.

Collaborative Filtering Recommendation Systems

Reducing Dueling Bandits to Cardinal Bandits

no code implementations14 May 2014 Nir Ailon, Thorsten Joachims, Zohar Karnin

We present algorithms for reducing the Dueling Bandits problem to the conventional (stochastic) Multi-Armed Bandits problem.

Multi-Armed Bandits

Volumetric Spanners: an Efficient Exploration Basis for Learning

no code implementations21 Dec 2013 Elad Hazan, Zohar Karnin, Raghu Mehka

Numerous machine learning problems require an exploration basis - a mechanism to explore the action space.

BIG-bench Machine Learning Efficient Exploration

Near-Optimal Entrywise Sampling for Data Matrices

no code implementations NeurIPS 2013 Dimitris Achlioptas, Zohar Karnin, Edo Liberty

We consider the problem of selecting non-zero entries of a matrix $A$ in order to produce a sparse sketch of it, $B$, that minimizes $\|A-B\|_2$.

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