Blog

Jan 18, 2017 · post

New Research on Probabilistic Programming

image

We’re excited to release the latest research from our machine intelligence R&D team! 

This report and prototype explore probabilistic programming, an emerging programming paradigm that makes it easier to construct and fit Bayesian inference models in code. It’s advanced statistics, simplified for data scientists looking to build models fast.

Bayesian inference has been popular in scientific research for a long time. The statistical technique allows us to encode expert knowledge into a model by stating prior beliefs about what we think our data looks like. These prior beliefs are then updated in light of new data, providing not one prediction, but a full distribution of likely answers with baked-in confidence rates. This allows us to asses the risk of our decisions with more nuance.

Bayesian methods lack widespread commercial use because they’re tough to implement. But probabilistic programming reduces what used to take months of thorny statistical sampling into an afternoon of work.

This will further expand the utility of machine learning. Bayesian models aren’t black boxes, a criterion for regulated industries like healthcare. Unlike deep learning networks, they don’t require large, clean data sets or large amounts of GPU processing power to deliver results. And they bridge human knowledge with data, which may lead to breakthroughs in areas as diverse as anomaly detection and music analysis. 

Our work on probabilistic programming includes two prototypes and a report that teaches you:

  • How Bayesian inference works and where it’s useful
  • Why probabilistic programming is becoming possible now
  • When to use probabilistic programming and what the code looks like
  • What tools and languages exist today and how they compare
  • Which vendors offer probabilistic programming products

Finally, as in all our research, we predict where this technology is going, and applications for which it will be useful in the next couple of years.

Probabilistic Real Estate Prototype

One powerful feature of probabilistic programming is the ability to build hierarchical models, which allow us to group observations together and learn from their similarities. This is practical in contexts like user segmentation: individual users often shares tastes with other users of the same sex, age group, or location, and hierarchical models provide more accurate predictions about individuals by leveraging learnings from the group.

image

We explored using probabilistic programming for hierarchical models in our Probabilistic Real Estate prototype. This prototype predicts future real estate prices across the New York City boroughs. It enables you to input your budget (say $1.6 million) and shows you the probability of finding properties in that price range across different neighborhoods and future time periods.

Hierarchical models helped make predictions in neighborhoods with sparse pricing data. In our model, we declared that apartments are in neighborhoods and neighborhoods are in boroughs; on average, apartments in one neighborhood are more similar to others in the same location than elsewhere. By modeling this way, we could learn about the West Village not only from the West Village, but also from the East Village and Brooklyn. That means, with little data about the West Village, we could use data from the East Village to fill in the gaps! 

Many companies suffer from imperfect, incomplete data. These types of inferences can be invaluable to improve predictions based on real-world dependencies.

Play around with the prototype! You’ll see how the color gradients give you an intuitive sense for what probability distributions look like in practice.

Read more

Newer
Jan 25, 2017 · interview
Older
Jan 11, 2017 · interview

Latest posts

Sep 22, 2021 · post

Automatic Summarization from TextRank to Transformers
by Melanie Beck · Automatic summarization is a task in which a machine distills a large amount of data into a subset (the summary) that retains the most relevant and important information from the whole. While traditionally applied to text, automatic summarization can include other formats such as images or audio. In this article we’ll cover the main approaches to automatic text summarization, talk about what makes for a good summary, and introduce Summarize. – a summarization prototype we built that showcases several automatic summarization techniques.
...read more
Sep 21, 2021 · post

Extractive Summarization with Sentence-BERT
by Victor Dibia · In extractive summarization, the task is to identify a subset of text (e.g., sentences) from a document that can then be assembled into a summary. Overall, we can treat extractive summarization as a recommendation problem. That is, given a query, recommend a set of sentences that are relevant. The query here is the document, relevance is a measure of whether a given sentence belongs in the document summary. How we go about obtaining this measure of relevance varies (a common dilemma for any recommendation system).
...read more
Sep 20, 2021 · post

How (and when) to enable early stopping for Gensim's Word2Vec
by Melanie Beck · The Gensim library is a staple of the NLP stack. While it primarily focuses on topic modeling and similarity for documents, it also supports several word embedding algorithms, including what is likely the best-known implementation of Word2Vec. Word embedding models like Word2Vec use unlabeled data to learn vector representations for each token in a corpus. These embeddings can then be used as features in myriad downstream tasks such as classification, clustering, or recommendation systems.
...read more
Jul 7, 2021 · post

Exploring Multi-Objective Hyperparameter Optimization
By Chris and Melanie. The machine learning life cycle is more than data + model = API. We know there is a wealth of subtlety and finesse involved in data cleaning and feature engineering. In the same vein, there is more to model-building than feeding data in and reading off a prediction. ML model building requires thoughtfulness both in terms of which metric to optimize for a given problem, and how best to optimize your model for that metric!
...read more
Jun 9, 2021 ·

Deep Metric Learning for Signature Verification
By Victor and Andrew. TLDR; This post provides an overview of metric learning loss functions (constrastive, triplet, quadruplet, and group loss), and results from applying contrastive and triplet loss to the task of signature verification. A complete list of the posts in this series is outlined below: Pretrained Models as Baselines for Signature Verification -- Part 1: Deep Learning for Automatic Offline Signature Verification: An Introduction Part 2: Pretrained Models as Baselines for Signature Verification Part 3: Deep Metric Learning for Signature Verification In our previous blog post, we discussed how pretrained models can serve as strong baselines for the task of signature verification.
...read more
May 27, 2021 · post

Pretrained Models as a Strong Baseline for Automatic Signature Verification
By Victor and Andrew. Figure 1. Baseline approach for automatic signature verification using pretrained models TLDR; This post describes how pretrained image classification models can be used as strong baselines for the task of signature verification. The full list of posts in the series is outlined below: Pretrained Models as Baselines for Signature Verification -- Part 1: Deep Learning for Automatic Offline Signature Verification: An Introduction Part 2: Pretrained Models as Baselines for Signature Verification Part 3: Deep Metric Learning for Signature Verification As discussed in our introductory blog post, offline signature verification is a biometric verification task that aims to discriminate between genuine and forged samples of handwritten signatures.
...read more

Popular posts

Oct 30, 2019 · newsletter
Exciting Applications of Graph Neural Networks
Nov 14, 2018 · post
Federated learning: distributed machine learning with data locality and privacy
Apr 10, 2018 · post
PyTorch for Recommenders 101
Oct 4, 2017 · post
First Look: Using Three.js for 2D Data Visualization
Aug 22, 2016 · whitepaper
Under the Hood of the Variational Autoencoder (in Prose and Code)
Feb 24, 2016 · post
"Hello world" in Keras (or, Scikit-learn versus Keras)

Reports

In-depth guides to specific machine learning capabilities
FF22

Inferring Concept Drift Without Labeled Data

Concept drift occurs when the statistical properties of a target domain change overtime 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.
Read the report  →
FF19

Session-based Recommender Systems

Being able to recommend an item of interest to a user (based on their past preferences) is a highly relevant problem in practice. 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. This report explores a simple, yet powerful, NLP-based approach (word2vec) to recommend a next item to a user. While NLP-based approaches are generally employed for linguistic tasks, here we exploit them to learn the structure induced by a user’s behavior or an item’s nature.
Read the report  →
FF18

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.
Read the report  →
FF16

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.
Read the report  →
FF15

Meta-Learning

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
Read the report  →
FF14

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.
Read the report  →
FF13

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.
Read the report  →
FF06-2020

Interpretability

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.
Read the report  →
FF12

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.
Read the report  →
FF10

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.
Read the report  →
FF09

Federated Learning

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.
Read the report  →
FF03-2019

Deep Learning for Image Analysis: 2019 Edition

Convolutional Neural Networks (CNN) excel at learning meaningful representations of features and concepts within images. These capabilities make CNNs extremely valuable for solving problems in domains such as medical imaging, autonomous driving, manufacturing, robotics, and urban planning. In this report, we show how to select the right deep learning models for image analysis tasks and techniques for debugging deep learning models.
Read the report  →
FF03

Deep Learning: Image Analysis

Deep learning, or highly-connected neural networks, offers fascinating new capabilities for image analysis. Using deep learning, computers can now learn to identify objects in images. This report explores the history and current state of the field, predicts future developments, and explains how to apply deep learning today.
Read the report  →

Prototypes

Machine learning prototypes and interactive notebooks
Library

NeuralQA

A usable library for question answering on large datasets.
https://neuralqa.fastforwardlabs.com
Notebook

Explain BERT for Question Answering Models

Tensorflow 2.0 notebook to explain and visualize a HuggingFace BERT for Question Answering model.
https://colab.research.google.com/drive/1tTiOgJ7xvy3sjfiFC9OozbjAX1ho8WN9?usp=sharing
Notebooks

NLP for Question Answering

Ongoing posts and code documenting the process of building a question answering model.
https://qa.fastforwardlabs.com
Notebook

Interpretability Revisited: SHAP and LIME

Explore how to use LIME and SHAP for interpretability.
https://colab.research.google.com/drive/1pjPzsw_uZew-Zcz646JTkRDhF2GkPk0N

About

Cloudera Fast Forward is an applied machine learning reseach group.
Cloudera   Blog   Twitter