Get up to speed on deep learning with this on-demand webinar

Overview

What is deep learning?

Deep learning is a subset of machine learning where neural networks — algorithms inspired by the human brain — learn from large amounts of data. Deep learning algorithms perform a task repeatedly and gradually improve the outcome through deep layers that enable progressive learning. It’s part of a broader family of machine learning methods based on neural networks.

Deep learning is making a big impact across industries. In life sciences, deep learning can be used for advanced image analysis, research, drug discovery, prediction of health problems and disease symptoms, and the acceleration of insights from genomic sequencing. In transportation, it can help autonomous vehicles adapt to changing conditions. It is also used to protect critical infrastructure and speed response.

Businesses often outsource the development of deep learning. However, it is better to keep the deep learning development work for use cases that are core to your business. These include fraud detection and recommendations, predictive maintenance and time-series data analysis, recommendation system optimization, customer relationship management, and predicting the clickthrough rate of online advertising.

Get started with deep learning using IBM Watson Studio® on IBM Cloud Pak® for Data as a Service.

Features

Experiment Builder

Initiate and monitor batch training experiments, compare cross-model performance in real time and focus on designing neural networks.

Distributed deep learning (DDL)

Enable popular open source frameworks such as TensorFlow, Caffe, Torch and Chainer to scale to multiple GPUs.

Handwritten digit recognition

Use a pretrained PyTorch model to predict handwritten numbers from images. Use REST APIs to submit training jobs, monitor status, and store and deploy models.

Visual recognition service

Use deep learning algorithms from IBM Watson Visual Recognition service to analyze images for scenes and objects. Work with images and datasets in a collaborative environment.

Image classification

Perform multiclass classification, preprocess and access images, and create visualizations to gain a better understanding of your models.

Language models

Use a notebook, Keras and TensorFlow to build a language model for text generation.

See deep learning in IBM Watson Studio

Deep learning experiment

Product screen shot showing where you define the metadata for a new deep learning experiment

Deep learning experiment

Run a deep learning experiment to create a training run for each definition.

Model definitions

Product screen shot showing where you add a model definition, including the name, training source code, framework and execution command

Model definitions

Define your model building code, execution command, a GPU and other metadata.

Resource plan

Product screen shot showing the resource plan for a project including an overview tab of GPU configurations, a user statistics tab and an active applications tab

Resource plan

Determine the GPU configurations in the resource plan.

Training progress

Product screen shot showing training status line graphs

Training progress

Monitor deep learning training.

GPU notebook

Product screen shot showing image classification within a GPU notebook

GPU notebook

Create a GPU environment definition and run your notebook at the time you create the notebook.

Use your favorite framework

Preinstalled and optimized for performance in IBM Watson Studio

TensorFlow logo
Keras logo
PyTorch logo

Get started with deep learning

Start executing your deep learning experiments on IBM Watson Studio.