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.

Fast Object Detection in Compressed Video

Object detection in videos has drawn increasing attention recently since it is more important in real scenarios. Most of the deep learning methods for video analysis use convolutional neural networks designed for image-wise parsing in a video stream. But they usually ignore the fact that a video is generally stored and transmitted in a compressed data format. In this paper, we propose a fast object detection model that incorporates light-weight motion-aided memory network (MMNet), which can be directly used for H.264 compressed video. MMNet has two major advantages: 1) For a group of successive pictures (GOP) in a compressed video stream, it runs the heavy computational network for I-frames, i.e. a few reference frames in videos, while a light-weight memory network is designed to generate features for prediction frames called P-frames; 2) Unlike establishing an additional network to explicitly model motion among frames, we directly take full advantage of both motion vectors and residual errors that are all encoded in a compressed video. Such signals maintain spatial variations and are freely available. To our best knowledge, the MMNet is the first work that explores a convolutional detector on a compressed video and a motion-based memory in order to achieve significant speedup. Our model is evaluated on the large-scale ImageNet VID dataset, and the results show that it is about 3x times faster than single image detector R-FCN and 10x times faster than high performance detectors like FGFA and MANet.

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.