LEARN
HOW
UBER
IS USING
MACHINE
LEARNING
@learn.machinelearning 01/10
Fare estimates
This Forecasting ML algorithm allows Uber to play with
the prices during peak hours to increase profit. Uber
calculates fares using real-time traffic data. It analyses
various external factors that could affect the fares, such
as public transport availability and how accessible these
public facilities are, etc.
@learn.machinelearning 02/10
Supply and demand chain
the Uber team can estimate the time and location of
demand by exploiting both the stored and the real-
time data coming from the users' app. Using these
estimates, the app notifies the drivers to meet the
demand requirements in a particular area.
@learn.machinelearning 03/10
Expected time arrival
Uber uses real-time traffic and GPS data and
Map APls to forecast the expected arrival time.
Uber AI partnered with Uber’s Maps team on a
project called DeepETA to develop a low-latency
deep neural network architecture for global ETA
prediction.
@learn.machinelearning 04/10
Route Optimization
Uber's ML system predicts the best routes and
recommend the most optimal ones to the drivers.
Through this process, Uber assists drivers to avert
crowding areas and enables speedy rides. It not only
makes customers pleased but also offers drivers extra
time to conduct additional rides.
@learn.machinelearning 05/10
Single-click communication
Leveraging Natural Language Processing and Machine
Learning technique to frame answers to common
messages. By this way, cab drivers could respond
effortlessly by just clicking on one of the suggested
responses.
@learn.machinelearning 06/10
Uber pool
Ml system can decide whom to pick first and drop first
on the basis of data accumulated from map. Uber Pool
also uses the stored data to find out the hidden
pattern and accordingly modify the prices to offer the
best services to its customers and, at the same time,
manage higher profits.
@learn.machinelearning 07/10
Self-Driving Cars
Uber’s self-driving car systems use ML for functions
like object detection, motion planning and efficient
distributed training of large models across a large
number of GPU machines.
@learn.machinelearning 08/10
Open source contribution
Ludwig: A toolbox that allows to train and test deep
learning models without the need to write code
Pyro: A Native Probabilistic Programming Language
Manifold: A Debugging and Interpretation Toolset for
Machine Learning Models
Plato: A Framework for Building Conversational Agents
at Scale
Horovod: Distributed training framework for
TensorFlow, Keras, PyTorch, and Apache MXNet
Petastorm: Petastorm library enables single machine
or distributed training and evaluation of deep learning
models
AresDB: A GPU-powered real-time analytics storage
and query engine
@learn.machinelearning 08/10
Turn on post notifications
Which company do you want
me to cover next?
Like Save
