Fast Forward Labs research now available without a subscription
Moving forward, all new reports will be publicly available and free to download. In addition, we will be providing access to updated versions of older reports over time, so check back often to explore available free research.
Text Style Transfer
The NLP task of text style transfer (TST) aims to automatically control the style attributes of a piece of text while preserving the content, which is an important consideration for making NLP more user-centric. In this report, we explore text style transfer through an applied use case — neutralizing subjectivity bias in free text. Along the way, we describe our sequence-to-sequence modeling approach leveraging HuggingFace Transformers, and present a set of custom, reference-free evaluation metrics for quantifying model performance. Finally, we conclude with a discussion of ethics centered around our prototype: Exploring Intelligent Writing Assistance.
Inferring Concept Drift Without Labeled Data
Concept drift occurs when the statistical properties of a target domain change over time causing model performance to degrade. Drift detection is generally achieved by monitoring a performance metric of interest and triggering a retraining pipeline when that metric falls below some designated threshold. However, this approach assumes ample labeled data is available at prediction time - an unrealistic constraint for many production systems. In this report, we explore various approaches for dealing with concept drift when labeled data is not readily accessible.
Exploring Multi-Objective Hyperparameter Optimization
We develop machine learning models against the “usual suspect” metrics like predictive accuracy, recall, and precision. However, these metrics are rarely truly all we care about. Production models must also satisfy physical requirements such as latency or memory footprint, or fairness constraints. Hyperparameter optimization becomes even more challenging when we have multiple metrics to optimize. Our latest research examines this “multi-objective” hyperparameter optimization scenario in detail.
Deep Learning for Automatic Offline Signature Verification
Handwritten signature verification aims to automatically discriminate between genuine and forged signatures, and is a particularly important challenge due to the ubiquity of handwritten signatures as a form of identification in legal, financial, and administrative domains. This research cycle explored the use of deep metric learning approaches - specifically siamese networks - combined with novel feature extraction methods to improve upon traditional techniques.
Session-Based Recommender Systems
Recommendation systems have become a cornerstone of modern life, spanning sectors that include online retail, music and video streaming, and even content publishing. These systems help us navigate the sheer volume of content on the internet, allowing us to discover what’s interesting or important to us. A key trend over the past few years has been session-based recommendation algorithms that provide recommendations solely based on a user’s interactions in an ongoing session, and which do not require the existence of user profiles or their entire historical preferences.
Few-Shot Text Classification
Text classification can be used for sentiment analysis, topic assignment, document identification, article recommendation, and more. While dozens of techniques now exist for this fundamental task, many of them require massive amounts of labeled data in order to be useful. Collecting annotations for your use case is typically one of the most costly parts of any machine learning application. In this report, we explore how latent text embeddings can be used with few (or even zero) training examples and provide insights into best practices for implementing this method.
Structural Time Series
Time series data is ubiquitous. This report examines generalized additive models, which give us a simple, flexible, and interpretable means for modeling time series by decomposing them into structural components. We look at the benefits and trade-offs of taking a curve-fitting approach to time series, and demonstrate its use via Facebook’s Prophet library on a demand forecasting problem.
In contrast to how humans learn, deep learning algorithms need vast amounts of data and compute and may yet struggle to generalize. Humans are successful in adapting quickly because they leverage their knowledge acquired from prior experience when faced with new problems. In this report, we explain how meta-learning can leverage previous knowledge acquired from data to solve novel tasks quickly and more efficiently during test time
Automated Question Answering
Automated question answering is a user-friendly way to extract information from data using natural language. Thanks to recent advances in natural language processing, question answering capabilities from unstructured text data have grown rapidly. This blog series offers a walk-through detailing the technical and practical aspects of building an end-to-end question answering system.
Causality for Machine Learning
The intersection of causal inference and machine learning is a rapidly expanding area of research that's already yielding capabilities to enable building more robust, reliable, and fair machine learning systems. This report offers an introduction to causal reasoning including causal graphs and invariant prediction and how to apply causal inference tools together with classic machine learning techniques in multiple use-cases.
Interpretability: 2020 Edition
Interpretability, or the ability to explain why and how a system makes a decision, can help us improve models, satisfy regulations, and build better products. Black-box techniques like deep learning have delivered breakthrough capabilities at the cost of interpretability. In this report, recently updated to include techniques like SHAP, we show how to make models interpretable without sacrificing their capabilities or accuracy.
Deep Learning for Anomaly Detection
From fraud detection to flagging abnormalities in imaging data, there are countless applications for automatic identification of abnormal data. This process can be challenging, especially when working with large, complex data. This report explores deep learning approaches (sequence models, VAEs, GANs) for anomaly detection, when to use them, performance benchmarks, and product possibilities.
Transfer Learning for Natural Language Processing
Natural language processing (NLP) technologies using deep learning can translate language, answer questions, and generate human-like text But these deep learning techniques require large, costly labeled datasets, expensive infrastructure, and scarce expertise. Transfer learning lifts these constraints by reusing and adapting a model’s understanding of language. Transfer learning is a good fit for any NLP application. In this report, we show how to use transfer learning to build high-performance NLP systems with minimal resources.
Learning with Limited Labeled Data
Being able to learn with limited labeled data relaxes the stringent labeled data requirement for supervised machine learning. This report focuses on active learning, a technique that relies on collaboration between machines and humans to label smartly. Active learning reduces the number of labeled examples required to train a model, saving time and money while obtaining comparable performance to models trained with much more data. With active learning, enterprises can leverage their large pool of unlabeled data to open up new product possibilities.
Federated Learning makes it possible to build machine learning systems without direct access to training data. The data remains in its original location, which helps to ensure privacy and reduces communication costs. Federated learning is a great fit for smartphones and edge hardware, healthcare and other privacy-sensitive use cases, and industrial applications such as predictive maintenance.
The internet has given us an avalanche of options for what to read, watch and buy. Because of this, recommendation algorithms, which find items that will interest a particular person, are more important than ever. In this report we explore recommendation systems that make use of the semantic content of items and users to deliver richer recommendations across multiple industries.
This report explores methods for extractive summarization, a capability that allows one to automatically summarize documents. This technique has a wealth of applications: from the ability to distill thousands of product reviews, extract the most important content from long news articles, or automatically cluster customer bios into personas.
Deep Learning for Image Analysis - 2019 Edition
Convolutional neural networks (CNNs or ConvNets) excel at learning meaningful representations of features and concepts within images, making CNNs valuable for solving problems in multiple domains, from medical imaging to manufacturing. In this report, we show how to select the right deep learning models for image analysis tasks and techniques for debugging deep learning models.
Probabilistic Methods for Realtime Streams
Since the days of analog computers built on cams and gears, we’ve been engineering systems around the flow of data and the critical calculations we must perform. While the philosophy of our designs has remained consistent, our engineering constraints are constantly evolving. In the past five years, we’ve seen the emergence of “big data,” or the ability to use commodity infrastructure to analyze very large data sets in a batch. We’re currently in the midst of a significant step forward in the tools, methods, and technologies available for working with real-time streams of data.
Updated versions of older reports will be available for free in the future, so check back often.
In this report, we focus on multi-task learning, a new approach to machine learning that allows algorithms to master tasks in parallel.
Here, we show how to use probabilistic programming and Bayesian inference to easily build tools that make better predictions for more effective decision making.
Natural language generation
In this report, we look at how machine systems can turn highly structured data into human language narrative.
Keep up with tomorrow
Sign up for our monthly newsletter and get the latest on advances in applied artificial intelligence, as well as company news and events.