Publications

We introduce a new Collaborative Causal Discovery problem, through which we model a common scenario in which we have multiple independent entities each with their own causal graph, and the goal is to simultaneously learn all these causal graphs. We study this problem without the causal sufficiency assumption, using Maximal Ancestral Graphs (MAG) to model the causal graphs, and assuming that we have the

Active learning is a commonly used technique to reduce the amount of labeled data necessary for supervised learning. In this paper, we focus on collection of labeled examples in a domain with large unlabeled dataset and extreme class imbalance. This scenario presents several challenges to Active learning. Traditional active learning strategies can face acute difficulty in locating minority class examples

Episodic training is a core ingredient of few-shot learning to train models on tasks with limited labelled data. Despite its success, episodic training remains largely understudied, prompting us to ask the question: what is the best way to sample episodes? In this paper, we first propose a method to approximate episode sampling distributions based on their difficulty. Building on this method, we perform

This article proposes novel rules for false discovery rate control (FDRC) geared towards online anomaly detection in time series. Online FDRC rules allow to control the properties of a sequence of statistical tests. In the context of anomaly detection, the null hypothesis is that an observation is normal and the alternative is that it is anomalous. FDRC rules allow users to target a lower bound on precision

Automatic Speech Recognition (ASR) systems form a key component of various products across industry. Many of these ASR systems rely on a complex Acoustic Model (AM) whose output is rescored by a domain-specific Language Model (LM). As we use ASR systems in new domains, the memory, maintenance and data-collection costs for these domain-specific LMs increase. Particularly, with advent of parameter-heavy Transformer

Neural ordinary differential equations describe how values change in time. This is the reason why they gained importance in modeling sequential data, especially when the observations are made at irregular intervals. In this paper we propose an alternative by directly modeling the solution curves — the flow of an ODE — with a neural network. This immediately eliminates the need for expensive numerical solvers

Controllable text generation concerns two fundamental tasks of wide applications, namely generating text of given attributes (i.e., attribute-conditional generation), and minimally editing existing text to possess desired attributes (i.e., text attribute transfer). Extensive prior work has largely studied the two problems separately, and developed different conditional models which, however, are prone to

In this work we discuss One-Shot Object Detection, a challenging task of detecting novel objects in a target scene using a single reference image called a query. To address this challenge we introduce SPOT (Surfacing POsitions using Transformers), a novel transformer based end-to-end architecture which uses synergy between the provided query and target images using a learnable Robust Feature Matching module

Learning the distribution of future trajectories conditioned on the past is a crucial problem for understanding multi-agent systems. This is challenging because humans make decisions based on complex social relations and personal intents, resulting in highly complex uncertainties over trajectories. To address this problem, we propose a conditional deep generative model that combines advances in graph neural

Companies rely on large-scale surveys, interviews, and focus groups to gauge customer sentiment about their products or programs, which contain free form text data rich in information. Researchers currently use a manual, time consuming processing which delays the time to get actionable insights. This paper presents a scalable solution where researchers can interact with a custom UI to annotate text data

Proxy-based metric learning losses are superior to pair-based losses due to their fast convergence and low training complexity. However, existing proxy-based losses focus on learning class-discriminative features while overlooking the commonalities shared across classes which are potentially useful in describing and matching samples. Moreover, they ignore the implicit hierarchy of categories in real-world

Label inference was recently introduced as the problem of reconstructing the ground truth labels of a private dataset from just the (possibly perturbed) cross entropy loss scores evaluated at carefully crafted prediction vectors. In this paper, we generalize this result to provide necessary and sufficient conditions under which label inference is possible from a broad class of loss functions. We show that

Text-to-Text (T2T) denoising-pretraining-finetuning (DPF) paradigms (e.g. BERT, BART, GPT) have achieved great success in a wide range of encoding and decoding tasks in NLP. However, little has been explored on data-to-data (D2D) and data-to-text (D2T) tasks using DPF paradigms. This work fills in the gap by investigating D2D and T2T denoising-pretraining for D2T tasks. D2D and T2T DPF paradigms can leverage

Building supervised targeted sentiment analysis models for a new target domain requires substantial annotation effort since most datasets for this task are domain-specific. Domain adaptation for this task has two dimensions: the nature of targets and the opinion words used to describe sentiment towards the target. We present a data sampling strategy informed by domain differences across these two dimensions

A challenge for target-based sentiment analysis is that most datasets are domain-specific and thus building supervised models for a new target domain requires substantial annotation effort. Domain adaptation for this task has two dimensions: the nature of the targets (e.g., entity types, properties associated with entities, or arbitrary spans) and the opinion words used to describe the sentiment towards