Goto

Collaborating Authors

When to Trust Your Model: Model-Based Policy Optimization

Neural Information Processing Systems

Designing effective model-based reinforcement learning algorithms is difficult because the ease of data generation must be weighed against the bias of model-generated data. In this paper, we study the role of model usage in policy optimization both theoretically and empirically. We first formulate and analyze a model-based reinforcement learning algorithm with a guarantee of monotonic improvement at each step. In practice, this analysis is overly pessimistic and suggests that real off-policy data is always preferable to model-generated on-policy data, but we show that an empirical estimate of model generalization can be incorporated into such analysis to justify model usage. Motivated by this analysis, we then demonstrate that a simple procedure of using short model-generated rollouts branched from real data has the benefits of more complicated model-based algorithms without the usual pitfalls.


How to build deep learning models with SAS

#artificialintelligence

SAS supports the creation of deep neural network models. Examples of these models include convolutional neural networks, recurrent neural networks, feedforward neural networks and autoencoder neural networks. Let's examine in more detail how SAS creates deep learning models using SAS Visual Data Mining and Machine Learning. SAS Visual Mining and Machine Learning takes advantage of SAS Cloud Analytic Services (CAS) to perform what are referred to as CAS actions. You use CAS actions to load data, transform data, compute statistics, perform analytics and create output.


How to build deep learning models with SAS

#artificialintelligence

SAS supports the creation of deep neural network models. Examples of these models include convolutional neural networks, recurrent neural networks, feedforward neural networks and autoencoder neural networks. Let's examine in more detail how SAS creates deep learning models using SAS Visual Data Mining and Machine Learning. SAS Visual Mining and Machine Learning takes advantage of SAS Cloud Analytic Services (CAS) to perform what are referred to as CAS actions. You use CAS actions to load data, transform data, compute statistics, perform analytics and create output.


Hey Alexa! Sorry I fooled you ...

#artificialintelligence

A human can likely tell the difference between a turtle and a rifle. For quite some time, a subset of computer science research has been dedicated to better understanding how machine-learning models handle these "adversarial" attacks, which are inputs deliberately created to trick or fool machine-learning algorithms. While much of this work has focused on speech and images, recently, a team from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) tested the boundaries of text. They came up with "TextFooler," a general framework that can successfully attack natural language processing (NLP) systems -- the types of systems that let us interact with our Siri and Alexa voice assistants -- and "fool" them into making the wrong predictions. One could imagine using TextFooler for many applications related to internet safety, such as email spam filtering, hate speech flagging, or "sensitive" political speech text detection -- which are all based on text classification models.


My Favorite Machine-Learning Models ALL Data-Scientists Should Know

#artificialintelligence

These are my subjective favorite models, and the ones that I find myself turning to most often. There is a tool for every job just as there is a model for every job, just as there are features for every job. A great understanding can come from a multitude of sources, but as a lifetime learner, never expect to get too comfortable with one model or another. I'm moderately guilty of this in some circumstances, because it can be comforting to return to what you know, but traveling to the dark depths of stack overflow is likely the greatest education you can get for Data Science. Among many other things, a strong knowledge of Machine Learning models will come soon after.