Nand Kishor is the Product Manager of House of Bots. After finishing his studies in computer science, he ideated & re-launched Real Estate Business Intelligence Tool, where he created one of the leading Business Intelligence Tool for property price analysis in 2012. He also writes, research and sharing knowledge about Artificial Intelligence (AI), Machine Learning (ML), Data Science, Big Data, Python Language etc... ...
Full BioNand Kishor is the Product Manager of House of Bots. After finishing his studies in computer science, he ideated & re-launched Real Estate Business Intelligence Tool, where he created one of the leading Business Intelligence Tool for property price analysis in 2012. He also writes, research and sharing knowledge about Artificial Intelligence (AI), Machine Learning (ML), Data Science, Big Data, Python Language etc...
3 Best Programming Languages For Internet of Things Development In 2018
879 days ago
Data science is the big draw in business schools
1052 days ago
7 Effective Methods for Fitting a Liner
1062 days ago
3 Thoughts on Why Deep Learning Works So Well
1062 days ago
3 million at risk from the rise of robots
1062 days ago
Top 10 Hot Artificial Intelligence (AI) Technologies
345669 views
2018 Data Science Interview Questions for Top Tech Companies
97710 views
Want to be a millionaire before you turn 25? Study artificial intelligence or machine learning
92646 views
Here's why so many data scientists are leaving their jobs
90264 views
Google announces scholarship program to train 1.3 lakh Indian developers in emerging technologies
70095 views
Applying Deep Learning to Real World Problems
The rise of artificial intelligence in recent years is grounded in the success of deep learning. Three major drivers caused the breakthrough of (deep) neural networks: the availability of huge amounts of training data, powerful computational infrastructure, and advances in academia. Thereby deep learning systems start to outperform not only classical methods, but also human benchmarks in various tasks like image classification or face recognition. This creates the potential for many disruptive new businesses leveraging deep learning to solve real-world problems.
At Berlin-based Merantix, we work on these new business cases in various industries (currently automotive, health, financial and advertising).
It easier than ever before to train a neural network. However, it is rarely the case that you can just take code from a tutorial and directly make it work for your application. Interestingly, many of the most important tweaks are barely discussed in the academic literature but at the same time critical to make your product work.
Therefore I thought it would be helpful for other people who plan to use deep learning in their business to understand some of these tweaks and tricks. In this blog post I want to share three key learnings, which helped us at Merantix when applying deep learning to real-world problems:
Learning I: The Value Of Pre-Training
Learning II: Caveats Of Real-World Label Distributions
Learning III: Understanding Black Box Models
A little disclaimer:
This is not a complete list and there are many other important tweaks.
Most of these learnings apply not only to deep learning but also to other machine learning algorithms.
All the learnings are industry-agnostic.
Most of the ideas in the post refer to supervised learning problems.
LEARNING I: THE VALUE OF PRE-TRAINING
In the academic world of machine learning, there is little focus on obtaining datasets. Instead, it is even the opposite: in order to compare deep learning techniques with other approaches and ensure that one method outperforms others, the standard procedure is to measure the performance on a standard dataset with the same evaluation procedure. However, in real-world scenarios, it is less about showing that your new algorithm squeezes out an extra 1% in performance compared to another method. Instead it is about building a robust system which solves the required task with sufficient accuracy. As for all machine learning systems, this requires labeled training from which the algorithm can learn from.
For many real-world problems it is unfortunately rather expensive to get well-labeled training data. To elaborate on this issue, letâ??s consider two cases:
1. Medical vision: if we want to build a system which detects lymph nodes in the human body in Computed Tomography (CT) images, we need annotated images where the lymph node is labeled. This is a rather time consuming task, as the images are in 3D and it is required to recognize very small structures. Assuming that a radiologist earns 100$/h and can carefully annotate 4 images per hour, this implies that we incur costs of 25$ per image or 250k$ for 10000 labeled images. Considering that we require several physicians to label the same image to ensure close to 100% diagnosis correctness, acquiring a dataset for the given medical task would easily exceed those 250k$.
2. Credit scoring: if we want to build a system that makes credit decisions, we need to know who is likely to default so we can train a machine learning system to recognize them beforehand. Unfortunately, you only know for sure if somebody defaults when it happens. Thus a naive strategy would be to give loans of say 10k$ to everyone. However, this means that every person that defaults will cost us 10k$. This puts a very expensive price tag on each labeled datapoint.
Obviously there are tricks to lower these costs, but the overall message is that labeled data for real-world problems can be expensive to obtain. Read More