arXiv subjects

Pavel Dmitriev

Publications and source records attributed to Pavel Dmitriev.

4 recordsLinked to original sources

arXiv2019

An Evaluation of Transfer Learning for Classifying Sales Engagement Emails at Large Scale

This paper conducts an empirical investigation to evaluate transfer learning for classifying sales engagement emails arising from digital sales engagement platforms. Given the complexity of content and context of sales engagement, lack of standardized large corpora and benchmarks, limited labeled examples and heterogenous context of intent, this real-world use case poses both a challenge and an opportunity for adopting a transfer learning approach. We propose an evaluation framework to assess a high performance transfer learning (HPTL) approach in three key areas in addition to commonly used accuracy metrics: 1) effective embeddings and pretrained language model usage, 2) minimum labeled samples requirement and 3) transfer learning implementation strategies. We use in-house sales engagement email samples as the experiment dataset, which includes over 3000 emails labeled as positive, objection, unsubscribe, or not-sure. We discuss our findings on evaluating BERT, ELMo, Flair and GloVe embeddings with both feature-based and fine-tuning approaches and their scalability on a GPU cluster with increasingly larger labeled samples. Our results show that fine-tuning of the BERT model outperforms with as few as 300 labeled samples, but underperforms with fewer than 300 labeled samples, relative to all the feature-based approaches using different embeddings.

cs.CL↗

arXiv2019

Trustworthy Experimentation Under Telemetry Loss

Failure to accurately measure the outcomes of an experiment can lead to bias and incorrect conclusions. Online controlled experiments (aka AB tests) are increasingly being used to make decisions to improve websites as well as mobile and desktop applications. We argue that loss of telemetry data (during upload or post-processing) can skew the results of experiments, leading to loss of statistical power and inaccurate or erroneous conclusions. By systematically investigating the causes of telemetry loss, we argue that it is not practical to entirely eliminate it. Consequently, experimentation systems need to be robust to its effects. Furthermore, we note that it is nontrivial to measure the absolute level of telemetry loss in an experimentation system. In this paper, we take a top-down approach towards solving this problem. We motivate the impact of loss qualitatively using experiments in real applications deployed at scale, and formalize the problem by presenting a theoretical breakdown of the bias introduced by loss. Based on this foundation, we present a general framework for quantitatively evaluating the impact of telemetry loss, and present two solutions to measure the absolute levels of loss. This framework is used by well-known applications at Microsoft, with millions of users and billions of sessions. These general principles can be adopted by any application to improve the overall trustworthiness of experimentation and data-driven decision making.

cs.OH↗

arXiv2017

Mid-infrared directional surface waves on a high aspect ratio nano-trench platform

Optical surface waves, highly localized modes bound to the surface of media, enable manipulation of light at nanoscale, thus impacting a wide range of areas in nanoscience. By applying metamaterials, artificially designed optical materials, as contacting media at the interface, we can significantly ameliorate surface wave propagation and even generate new types of waves. Here, we demonstrate that high aspect ratio (1:20) grating structures with plasmonic lamellas in deep nanoscale trenches function as a versatile platform supporting both surface and volume infrared waves. The surface waves exhibit a unique combination of properties, such as directionality, broadband existence (from 4 μm to at least 14 μm and beyond) and high localization, making them an attractive tool for effective control of light in an extended range of infrared frequencies.

physics.optics↗

arXiv2006

Mining Generalized Graph Patterns based on User Examples

There has been a lot of recent interest in mining patterns from graphs. Often, the exact structure of the patterns of interest is not known. This happens, for example, when molecular structures are mined to discover fragments useful as features in chemical compound classification task, or when web sites are mined to discover sets of web pages representing logical documents. Such patterns are often generated from a few small subgraphs (cores), according to certain generalization rules (GRs). We call such patterns "generalized patterns"(GPs). While being structurally different, GPs often perform the same function in the network. Previously proposed approaches to mining GPs either assumed that the cores and the GRs are given, or that all interesting GPs are frequent. These are strong assumptions, which often do not hold in practical applications. In this paper, we propose an approach to mining GPs that is free from the above assumptions. Given a small number of GPs selected by the user, our algorithm discovers all GPs similar to the user examples. First, a machine learning-style approach is used to find the cores. Second, generalizations of the cores in the graph are computed to identify GPs. Evaluation on synthetic data, generated using real cores and GRs from biological and web domains, demonstrates effectiveness of our approach.

cs.DS↗