February 6, 2020

348 words 2 mins read

Deep learning coming to the tire industry: Warehouse staffing with RNN-LSTMs and pricing optimizations with DNNs

Deep learning coming to the tire industry: Warehouse staffing with RNN-LSTMs and pricing optimizations with DNNs

Deep learning has been a sweeping revolution in the world of AI and machine learning. But sometimes traditional industries can be left behind. Alex Liang details two solutions where deep learning is used: a warehouse staffing solution where LSTM RNNs are used for staffing level forecasting and a pricing recommendation solution where DNNs were used for data clustering and demand modeling.
Talk Title Deep learning coming to the tire industry: Warehouse staffing with RNN-LSTMs and pricing optimizations with DNNs
Speakers Alex (Tianchu) Liang (American Tire Distributors)
Conference O’Reilly Artificial Intelligence Conference
Conf Tag Put AI to Work
Location San Jose, California
Date September 10-12, 2019
URL Talk Page
Slides Talk Slides
Video

Deep learning has been a sweeping revolution in the current world of AI and machine learning. It uses convolutional neural networks (CNNs) to help Teslas see the road properly; it uses reinforcement deep learning to help SpaceX land rockets automatically, and it uses recurrent neural networks (RNNs) to make machines translate better. The list goes on and on. But traditional industries may not see how this new, hot technology can help them. Alex Liang details how the data science team at American Tire Distributors (ATD) uses machine learning solutions to rejuvenate the company. He used LSTM RNN models ensembled with fbProphet to generate staffing-level forecasts and further optimized with CVXPY for maximum optimality of staffing schedules. He also implemented deep neural nets as part of a pricing optimization pipeline, where DNNs are used for clustering as well as product demand modeling. The warehouse solution is now being used every day across the entire US in 140 distribution centers to cost-effectively staff more than 2,000 people daily and is on track to realize ~10% in labor cost savings and the pricing solutions are now being fully productionalized into the revenue management process, automating and optimizing pricing decisions for the majority of products. He outlines the overall business problem context, initial machine learning prototyping, resolving challenges in data and compute, and application automation. You’ll leave with key takeaways in developing this solution, including both technical and business lessons.

recurrent neural network automation automating management cnn lstm data science ai network forecast deep learning machine learning pipeline cluster neural network optimization
comments powered by Disqus
Dilated neural networks for time series forecasting

Dilated neural networks for time series forecasting

December 25, 2019

Dilated neural networks are a class of recently developed neural networks that achieve promising results in time series forecasting. Chenhui Hu discusses representative network architectures of dilated neural networks and demonstrates their advantages in terms of training efficiency and forecast accuracy by applying them to solve sales forecasting and financial time series forecasting problems.
Long-term real-time network traffic flow prediction using LSTM recurrent neural network

Long-term real-time network traffic flow prediction using LSTM recurrent neural network

February 4, 2020

Real-time traffic volume prediction is vital in proactive network management, and many forecasting models have been proposed to address this. However, most are unable to fully use the information in traffic data to generate efficient and accurate traffic predictions for a longer term. Wei Cai explores predicting multistep, real-time traffic volume using many-to-one LSTM and many-to-many LSTM.
Deep learning methods for natural language processing

Deep learning methods for natural language processing

January 1, 2020

Garrett Hoffman walks you through deep learning methods for natural language processing and natural language understanding tasks, using a live example in Python and TensorFlow with StockTwits data. Methods include word2vec, recurrent neural networks and variants (LSTM, GRU), and convolutional neural networks.
Industrialized capsule networks for text analytics

Industrialized capsule networks for text analytics

December 30, 2019

Vijay Agneeswaran and Abhishek Kumar offer an overview of capsule networks and explain how they help in handling spatial relationships between objects in an image. They also show how to apply them to text analytics. Vijay and Abhishek then explore an implementation of a recurrent capsule network and benchmark the RCN with capsule networks with dynamic routing on text analytics tasks.
From inception to insight: Accelerating AI productivity with GPUs (sponsored by Dell Technologies)

From inception to insight: Accelerating AI productivity with GPUs (sponsored by Dell Technologies)

February 5, 2020

Data scientists and machine learning engineers need the flexibility to work in multiple environments without wasting precious time configuring hardware and software and modifying code. Ramesh Radhakrishnan and John Zedlewski walk you through deploying a simple set of technologies for executing end-to-end pipelines entirely on GPUs.
LSTM-based time series anomaly detection using Analytics Zoo for Spark and BigDL

LSTM-based time series anomaly detection using Analytics Zoo for Spark and BigDL

January 8, 2020

Collecting and processing massive time series data (e.g., logs, sensor readings, etc.) and detecting the anomalies in real time is critical for many emerging smart systems, such as industrial, manufacturing, AIOps, and the IoT. Guoqiong Song explains how to detect anomalies in time series data using Analytics Zoo and BigDL at scale on a standard Spark cluster.