Posts

How can AI and Machine learning impact healthcare industry?

Healthcare industry is a recession-proof one. Even in times of economic meltdown and financial distress, the healthcare industry can hold its own because mankind will always need healthcare. In fact, during the Great Depression in the US, when the economy was facing a severe slowdown, the healthcare industry expanded, adding 852,000 jobs.

Healthcare AI in the US is slated to reach $6.6 billion in value by 2021.

From clinical trials to new drug research & development, and from innovative medical devices to technology like nanoparticles, AI, and ML has touched every point and has the power to transform them completely.

In fact, according to a study by Accenture, AI applications in healthcare can result in global savings to the tune of $150 billion by 2026.

The possibilities are endless, and the results unthinkable if AI can be properly used.

Here are some of the ways AI and ML can impact the healthcare industry:

1. Solving the Iron Triangle

A problem that has plagued the world for many years the triangle aims to solve a fundamental healthcare problem: that of good quality, accessible treatment at low cost.

Providing all three at the same time is a major challenge in healthcare, as the cost of healthcare is usually high. Here, trying to improve one factor harms another.

But AI can solve this problem in the near future without breaking the triangle, by improving the current healthcare cost-structure. The key to it is AI, and smart machines, that the patient can use for self-treatment for the majority of times, cutting down treatment costs drastically, by reducing human contact and improving quality of life.

2. Diagnostics and Imaging

The US FDA has drastically increased investment on AI in radiology and diagnostics. And it’s not without reason.

The IDx-DR became the first AI system cleared by the US FDA to provide diagnostic decisions. It was a breakthrough discovery to detect early mild diabetic retinopathy. The device was accurate 87.5% of the times, and also detected patients who didn’t have the condition, correctly up to 89.5% of times.

The US FDA also permitted marketing of the Viz.AI a type of clinical decision support system designed to analyze CT scan results to identify possibilities of a stroke in the patients and send the results to a specialist to identify any block.

In fact, diagnostics is fast becoming one of the significant drivers of AI investment in healthcare.

These advances can impact the healthcare industry in a novel way. As more and more devices become AI-enabled, the landscape of healthcare delivery will change.

3. Early screening 

Early screening in case of most diseases can drastically improve the mortality rates of patients and cut down treatment costs by over 50%.

Let’s take the example of colorectal cancer.

The 5-year survival for Stage 1 CRC is around 90%, as compared to only 10% for Stage 4.

Early detection of CRC can be ideally treated with a minimally invasive endoscopy at a low cost of less than $5,000 per year. However, in the case of late-stage CRC, it requires multidisciplinary treatment with multiple surgeries, chemotherapy, and radiation, skyrocketing the costs.

And that is why early detection is essential, and that’s exactly what AI can do. There are already apps on the market that are doing this. For example, Autism & Beyond is a revolutionary app that leveraged the power of Apple’s ResearchKit to gather videos of children and detect their preference for the development of autism, using AI software.

AI used for early screening can save billions of taxpayer dollars of taxpayer money every year, and reduce out of pocket expenditure in the US drastically.

4. Drug research & development

According to the California Biomedical Research Association, it takes around 12 years for a drug to be conceived in the laboratory and go to the patient.

Only 1 out of 5000 drugs that are selected for pre-clinical testing are then used for human testing, and only 20% of them make it to the market for human use.

(image)

The cost to develop a new drug now is more than $2.5 billion.

It is only recently that AI is being used in drug research and discovery. The power of AI can be leveraged to streamline the drug discovery and drug repurposing processes. It can identify patients best suited to the trial, can identify patients in the most need for new medications and can predict any side-effects and idiosyncrasies beforehand.

All of these, for a start, can lead to much safer clinical trials with no unwanted drug reactions.

And then, there is the question of lowering costs. In fact, a study by Carnegie Mellon and a German university estimated that AI could lower drug discovery costs by as much as 70%.

This, in turn, will be transferred to patients in the form of lower drug prices, which will increase accessibility to better medications for patients and improve population health in general.

5. Surgery 

AI-enabled robotic-assisted surgeries are taking over the US. They are increasingly being used to reduce surgeon variations and improve quality.

‘Artificial intelligence can help surgeons perform better’ quotes Dr. John Birkmeyer, a chief clinical officer at Sound Physicians.

Advanced analytics and machine learning techniques are being used concomitantly used to unleash critical insights from the billions of data elements associated with robotic-assisted surgery. If used properly, this can help overcome attendant inefficiencies and improve patient health outcomes.

Artificial intelligence helps surgeons make better clinical decisions in real-time during surgery, and helps them understand the dynamics of the patient, especially during complex operations. It also reduces the length of stay of patients by 21%.

This is ultimately reflected in the patient’s post-operative care and long-term health. It also prevents patient readmissions, saving millions of dollars annually.

A study involving 379 orthopedic patients found out that AI-assisted robotic surgery resulted in five times fewer complications as compared to surgeons working alone.

According to Accenture, AI-assisted robotic surgery could save the US healthcare industry $40 billion annually, by 2026.

6. AI-assisted virtual nurses

AI-assisted virtual nurses could well end up saving the US healthcare industry $20 billion annually, by 2026.

They are available 24/7 to answer any patient queries, monitor patients, and guide them in any way they might want.

Currently, they act as a bridge for information exchange between care providers (doctors) and care receivers (patients), to decide what medications to start, the current health status, the most recent test results, and many other things.

It can save the patient many physical appointments with doctors, and also prevent high hospital readmission rates through simple, engaging, and intelligent care.

Care Angel is one of the finest virtual nurses around. Apart from all of the above, it can also provide wellness checks through voice and AI.

Wrap-Up 

AI and ML in healthcare are still at its infancy. Adoption at a large-scale is missing as of yet. To be successful in the healthcare domain, AI and ML need the endorsement of healthcare providers like physicians and nurses.

However, considerable investment is being made in AI in healthcare, and its increasing at a good rate.

AI in healthcare is currently aimed at improving patient outcomes, taking care of the interests of various stakeholders involved, increasing accessibility, and reducing healthcare costs.

In the near future, however, AI and ML, along with technologies like Data Science will take up a much more holistic role to drive healthcare forward.

Visual Question Answering with Keras – Part 2: Making Computers Intelligent to answer from images

Making Computers Intelligent to answer from images

This is my second blog on Visual Question Answering, in the last blog, I have introduced to VQA, available datasets and some of the real-life applications of VQA. If you have not gone through then I would highly recommend you to go through it. Click here for more details about it.

In this blog post, I will walk through the implementation of VQA in Keras.

You can download the dataset from here: https://visualqa.org/index.html. All my experiments were performed with VQA v2 and I have used a very tiny subset of entire dataset i.e all samples for training and testing from the validation set.

Table of contents:

  1. Preprocessing Data
  2. Process overview for VQA
  3. Data Preprocessing – Images
  4. Data Preprocessing through the spaCy library- Questions
  5. Model Architecture
  6. Defining model parameters
  7. Evaluating the model
  8. Final Thought
  9. References

NOTE: The purpose of this blog is not to get the state-of-art performance on VQA. But the idea is to get familiar with the concept. All my experiments were performed with the validation set only.

Full code on my Github here.


1. Preprocessing Data:

If you have downloaded the dataset then the question and answers (called as annotations) are in JSON format. I have provided the code to extract the questions, annotations and other useful information in my Github repository. All extracted information is stored in .txt file format. After executing code the preprocessing directory will have the following structure.

All text files will be used for training.

 

2. Process overview for VQA:

As we have discussed in previous post visual question answering is broken down into 2 broad-spectrum i.e. vision and text.  I will represent the Neural Network approach to this problem using the Convolutional Neural Network (for image data) and Recurrent Neural Network(for text data). 

If you are not familiar with RNN (more precisely LSTM) then I would highly recommend you to go through Colah’s blog and Andrej Karpathy blog. The concepts discussed in this blogs are extensively used in my post.

The main idea is to get features for images from CNN and features for the text from RNN and finally combine them to generate the answer by passing them through some fully connected layers. The below figure shows the same idea.

 

I have used VGG-16 to extract the features from the image and LSTM layers to extract the features from questions and combining them to get the answer.

3. Data Preprocessing – Images:

Images are nothing but one of the input to our model. But as you already may know that before feeding images to the model we need to convert into the fixed-size vector.

So we need to convert every image into a fixed-size vector then it can be fed to the neural network. For this, we will use the VGG-16 pretrained model. VGG-16 model architecture is trained on millions on the Imagenet dataset to classify the image into one of 1000 classes. Here our task is not to classify the image but to get the bottleneck features from the second last layer.

Hence after removing the softmax layer, we get a 4096-dimensional vector representation (bottleneck features) for each image.

Image Source: https://www.cs.toronto.edu/~frossard/post/vgg16/

 

For the VQA dataset, the images are from the COCO dataset and each image has unique id associated with it. All these images are passed through the VGG-16 architecture and their vector representation is stored in the “.mat” file along with id. So in actual, we need not have to implement VGG-16 architecture instead we just do look up into file with the id of the image at hand and we will get a 4096-dimensional vector representation for the image.

4. Data Preprocessing through the spaCy library- Questions:

spaCy is a free, open-source library for advanced Natural Language Processing (NLP) in Python. As we have converted images into a fixed 4096-dimensional vector we also need to convert questions into a fixed-size vector representation. For installing spaCy click here

You might know that for training word embeddings in Keras we have a layer called an Embedding layer which takes a word and embeds it into a higher dimensional vector representation. But by using the spaCy library we do not have to train the get the vector representation in higher dimensions.

 

This model is actually trained on billions of tokens of the large corpus. So we just need to call the vector method of spaCy class and will get vector representation for word.

After fitting, the vector method on tokens of each question will get the 300-dimensional fixed representation for each word.

5. Model Architecture:

In our problem the input consists of two parts i.e an image vector, and a question, we cannot use the Sequential API of the Keras library. For this reason, we use the Functional API which allows us to create multiple models and finally merge models.

The below picture shows the high-level architecture idea of submodules of neural network.

After concatenating the 2 different models the summary will look like the following.

The below plot helps us to visualize neural network architecture and to understand the two types of input:

 

6. Defining model parameters:

The hyperparameters that we are going to use for our model is defined as follows:

If you know what this parameter means then you can play around it and can get better results.

Time Taken: I used the GPU on https://colab.research.google.com and hence it took me approximately 2 hours to train the model for 5 epochs. However, if you train it on a PC without GPU, it could take more time depending on the configuration of your machine.

7. Evaluating the model:

Since I have used the very small dataset for performing these experiments I am not able to get very good accuracy. The below code will calculate the accuracy of the model.

 

Since I have trained a model multiple times with different parameters you will not get the same accuracy as me. If you want you can directly download mode.h5 file from my google drive.

 

8. Final Thoughts:

One of the interesting thing about VQA is that it a completely new field. So there is absolutely no end to what you can do to solve this problem. Below are some tips while replicating the code.

  1. Start with a very small subset of data: When you start implementing I suggest you start with a very small amount of data. Because once you are ready with the whole setup then you can scale it any time.
  2. Understand the code: Understanding code line by line is very much helpful to match your theoretical knowledge. So for that, I suggest you can take very few samples(maybe 20 or less) and run a small chunk (2 to 3 lines) of code to get the functionality of each part.
  3. Be patient: One of the mistakes that I did while starting with this project was to do everything at one go. If you get some error while replicating code spend 4 to 5 days harder on that. Even after that if you won’t able to solve, I would suggest you resume after a break of 1 or 2 days. 

VQA is the intersection of NLP and CV and hopefully, this project will give you a better understanding (more precisely practically) with most of the deep learning concepts.

If you want to improve the performance of the model below are few tips you can try:

  1. Use larger datasets
  2. Try Building more complex models like Attention, etc
  3. Try using other pre-trained word embeddings like Glove 
  4. Try using a different architecture 
  5. Do more hyperparameter tuning

The list is endless and it goes on.

In the blog, I have not provided the complete code you can get it from my Github repository.

9. References:

  1. https://blog.floydhub.com/asking-questions-to-images-with-deep-learning/
  2. https://tryolabs.com/blog/2018/03/01/introduction-to-visual-question-answering/
  3. https://github.com/sominwadhwa/vqamd_floyd

Accelerate your AI Skills Today: A Million Dollar Job!

The skyrocketing salaries ($1m per year) of AI engineers is not a hype. It is the fact of current corporate world, where you will witness a shift that is inevitable.

We’ve already set our feet at the edge of the technological revolution. A revolution that is at the verge of altering the way we live and work. As the fact suggests, humanity has fundamentally developed human production in three revolutions, and we’re now entering the fourth revolution. In its scope, the fourth revolution projects a transformation that is unlike anything we humans have ever experienced.

  • The first revolution had the world transformed from rural to urban
  • the emergence of mass production in the second revolution
  • third introduced the digital revolution
  • The fourth industrial revolution is anxious to integrate technologies into our lives.

And all thanks to artificial intelligence (AI). An advanced technology that surrounds us, from virtual assistants to software that translates to self-driving cars.

The rise of AI at an exponential rate has disrupted almost every industry. So much so that AI is being rated as one-million-dollar profession.

Did this grab your attention? It did?

Now, what if we were to tell you that the salary compensation for AI experts has grown dramatically. AI and machine learning are fields that have a mountain of demand in the tech industry today but has sparse supply.

AI field is growing at a quicker pace and salaries are skyrocketing! Read it for yourself to know what AI experts, AI researchers and any other AI talent are commanding today.

  • A top-class AI research laboratory, OpenAI says that techies in the AI field are projected to earn a salary compensation ranging between $300 to $500k for fresh graduates. However, expert professionals could earn anywhere up to $1m.
  • Whopping salary package of above 100 million yen that amounts to $1m is being offered to AI geniuses by a Japanese firm, Start Today. A firm that operates a fashion shopping website named Zozotown.

Does this leave you with a question – Is this a right opportunity for you to jump in the field and make hay while the sun is shining? 

And the answer to this question is – yes, it is the right opportunity for any developer seeking a role in the AI industry. It can be your chance to bridge the skill shortage in the AI field either by upskilling or reskilling yourself in the field of AI.

There are a wide varieties of roles available for an AI enthusiast like you. And certain areas are like AI Engineers and AI Researchers are high in demand, as there are not many professionals who have robust AI knowledge.

According to a job report, “The Future of Jobs 2018,” a prediction was made suggesting that machines and algorithms will create around 133 million new job roles by 2022.

AI and machine learning will dominate the tech world. The World Economic Forum says that several sectors have started embracing AI and machine learning to tackle challenges in certain fields such as advertising, supply chain, manufacturing, smart cities, drones, and cybersecurity.

Unraveling the AI realm

From chatbots to financial planners, AI is impacting the way businesses function on a day-today basis. AI makes the work simpler, as it provides variables, which makes the work more streamlined.

Alright! You know that

  • the demand for AI professionals is rising exponentially and that there is just a trickle of supply
  • the AI professionals are demanding skyrocketing salaries

However, beyond that how much more do you know about AI?

Considering the fact that our lives have already been touched by AI (think Alexa, and Siri), it is just a matter of time when AI will become an indispensable part of our lives.

As Gartner predicts that 2020 will be an important year for business growth in AI. Thus, it is possible to witness significant sparks for employment growth. Though AI predicts to diminish 1.8 million jobs, it is also said to replace it with 2.3 million jobs that will be created. As we look forward to stepping into 2020, AI-related job roles are set to make positive progress of achieving 2 million net-new employments by 2025.

With AI promising to score fat paychecks that would reach millions, AI experts are struggling to find new ways to pick up nouveau skills. However, one of the biggest impacts that affect the job market today is the scarcity of talent in this field.

The best way to stay relevant and employable in AI is probably by “reskilling,” and “upskilling.” And  AI certifications is considered ideal for those in the current workforce.

Looking to upskill yourself – here’s how you can become an AI engineer today.

Top three ways to enhance your artificial intelligence career:

  1. Acquire skills in Statistics and Machine Learning: If you’re getting into the field of machine learning, it is crucial that you have in-depth knowledge of statistics. Statistics is considered a prerequisite to the ML field. Both the fields are tightly related. Machine learning models are created to make accurate predictions while statistical models do the job of interpreting the relationship between variables. Many ML techniques heavily rely on the theory obtained through statistics. Thus, having extensive knowledge in statistics help initiate the first step towards an AI career.
  2. Online certification programs in AI skills: Opting for AI certifications will boost your credibility amongst potential employers. Certifications will also enhance your earning potential and increase your marketability. If you’re looking for a change and to be a part of something impactful; join the AI bandwagon. The IT industry is growing at breakneck speed; it is now that businesses are realizing how important it is to hire professionals with certain skillsets. Specifically, those who are certified in AI are becoming sought after in the job market.
  3. Hands-on experience: There’s a vast difference in theory and practical knowledge. One needs to familiarize themselves with the latest tools and technologies used by the industry. This is possible only if the individual is willing to work on projects and build things from scratch.

Despite all the promises, AI does prove to be a threat to job holders, if they don’t upskill or reskill themselves. The upcoming AI revolution will definitely disrupt the way we work, however, it will leave room for humans to perform more creative jobs in the future corporate world.

So a word of advice is to be prepared and stay future ready.

Visual Question Answering with Keras – Part 1

This is Part I of II of the Article Series Visual Question Answering with Keras

Making Computers Intelligent to answer from images

If we look closer in the history of Artificial Intelligence (AI), the Deep Learning has gained more popularity in the recent years and has achieved the human-level performance in the tasks such as Speech Recognition, Image Classification, Object Detection, Machine Translation and so on. However, as humans, not only we but also a five-year child can normally perform these tasks without much inconvenience. But the development of such systems with these capabilities has always considered an ambitious goal for the researchers as well as for developers.

In this series of blog posts, I will cover an introduction to something called VQA (Visual Question Answering), its available datasets, the Neural Network approach for VQA and its implementation in Keras and the applications of this challenging problem in real life. 

Table of Contents:

1 Introduction

2 What is exactly Visual Question Answering?

3 Prerequisites

4 Datasets available for VQA

4.1 DAQUAR Dataset

4.2 CLEVR Dataset

4.3 FigureQA Dataset

4.4 VQA Dataset

5 Real-life applications of VQA

6 Conclusion

 

  1. Introduction:

Let’s say you are given a below picture along with one question. Can you answer it?

I expect confidently you all say it is the Kitchen without much inconvenience which is also the right answer. Even a five-year child who just started to learn things might answer this question correctly.

Alright, but can you write a computer program for such type of task that takes image and question about the image as an input and gives us answer as output?

Before the development of the Deep Neural Network, this problem was considered as one of the difficult, inconceivable and challenging problem for the AI researcher’s community. However, due to the recent advancement of Deep Learning the systems are capable of answering these questions with the promising result if we have a required dataset.

Now I hope you have got at least some intuition of a problem that we are going to discuss in this series of blog posts. Let’s try to formalize the problem in the below section.

  1. What is exactly Visual Question Answering?:

We can define, “Visual Question Answering(VQA) is a system that takes an image and natural language question about the image as an input and generates natural language answer as an output.”

VQA is a research area that requires an understanding of vision(Computer Vision)  as well as text(NLP). The main beauty of VQA is that the reasoning part is performed in the context of the image. So if we have an image with the corresponding question then the system must able to understand the image well in order to generate an appropriate answer. For example, if the question is the number of persons then the system must able to detect faces of the persons. To answer the color of the horse the system need to detect the objects in the image. Many of these common problems such as face detection, object detection, binary object classification(yes or no), etc. have been solved in the field of Computer Vision with good results.

To summarize a good VQA system must be able to address the typical problems of CV as well as NLP.

To get a better feel of VQA you can try online VQA demo by CloudCV. You just go to this link and try uploading the picture you want and ask the related question to the picture, the system will generate the answer to it.

 

  1. Prerequisites:

In the next post, I will walk you through the code for this problem using Keras. So I assume that you are familiar with:

  1. Fundamental concepts of Machine Learning
  2. Multi-Layered Perceptron
  3. Convolutional Neural Network
  4. Recurrent Neural Network (especially LSTM)
  5. Gradient Descent and Backpropagation
  6. Transfer Learning
  7. Hyperparameter Optimization
  8. Python and Keras syntax
  1. Datasets available for VQA:

As you know problems related to the CV or NLP the availability of the dataset is the key to solve the problem. The complex problems like VQA, the dataset must cover all possibilities of questions answers in real-world scenarios. In this section, I will cover some of the datasets available for VQA.

4.1 DAQUAR Dataset:

The DAQUAR dataset is the first dataset for VQA that contains only indoor scenes. It shows the accuracy of 50.2% on the human baseline. It contains images from the NYU_Depth dataset.

Example of DAQUAR dataset

Example of DAQUAR dataset

The main disadvantage of DAQUAR is the size of the dataset is very small to capture all possible indoor scenes.

4.2 CLEVR Dataset:

The CLEVR Dataset from Stanford contains the questions about the object of a different type, colors, shapes, sizes, and material.

It has

  • A training set of 70,000 images and 699,989 questions
  • A validation set of 15,000 images and 149,991 questions
  • A test set of 15,000 images and 14,988 questions

Image Source: https://cs.stanford.edu/people/jcjohns/clevr/?source=post_page

 

4.3 FigureQA Dataset:

FigureQA Dataset contains questions about the bar graphs, line plots, and pie charts. It has 1,327,368 questions for 100,000 images in the training set.

4.4 VQA Dataset:

As comapred to all datasets that we have seen so far VQA dataset is relatively larger. The VQA dataset contains open ended as well as multiple choice questions. VQA v2 dataset contains:

  • 82,783 training images from COCO (common objects in context) dataset
  • 40, 504 validation images and 81,434 validation images
  • 443,757 question-answer pairs for training images
  • 214,354 question-answer pairs for validation images.

As you might expect this dataset is very huge and contains 12.6 GB of training images only. I have used this dataset in the next post but a very small subset of it.

This dataset also contains abstract cartoon images. Each image has 3 questions and each question has 10 multiple choice answers.

  1. Real-life applications of VQA:

There are many applications of VQA. One of the famous applications is to help visually impaired people and blind peoples. In 2016, Microsoft has released the “Seeing AI” app for visually impaired people to describe the surrounding environment around them. You can watch this video for the prototype of the Seeing AI app.

Another application could be on social media or e-commerce sites. VQA can be also used for educational purposes.

  1. Conclusion:

I hope this explanation will give you a good idea of Visual Question Answering. In the next blog post, I will walk you through the code in Keras.

If you like my explanations, do provide some feedback, comments, etc. and stay tuned for the next post.

Understanding Dropout and implementing it on MNIST dataset

Over-fitting is a major problem in deep learning and a plethora of techniques have been introduced to prevent it. One of the most effective one is called “dropout”.  Let’s use the analogy of a person going to gym for understanding this. Let’s say the person going to gym mostly uses his dominant arm, say his right arm to pick up weights. After some time, he notices that his dominant arm is developing a large muscle, but not the other arm. So, what can he do? Obviously, he needs to involve both his arms while training. Sometimes he should stop using his right arm, and use the left arm to lift weights and vice versa.

Something like this happens commonly in neural networks. Sometime one part of the network has very large weights and ends up dominating the training. While other part of the network remains weak and does not really play a role in the training. So, what dropout does to solve this problem, is it randomly shuts off some nodes and stop the gradients flowing through it. So, our forward and back propagation happen without those nodes. In that case the rest of the nodes need to pick up the slack and be more active in the training. We define a probability of the nodes getting dropped. For example, P=0.5 means there is a 50% chance a node will be dropped.

Figure 1 demonstrates the dropout technique, taken from the original research paper.

Dropout in a neuronal Net

Our network can never rely on any given node because it can be squashed at any given time. Hence the network is forced to learn redundant representation for everything to make sure at least some of the information remains. Redundant representation leads our network to be more robust. It also acts as ensemble of many networks, since at every epoch random nodes are dropped, each time our network will be different. Ensemble of different networks perform better than a single network since they capture more randomness. Please note, only non-output nodes are dropped.

Let’s, look at the python code to implement dropout in a neural network:

 

Künstliche Intelligenz und Vorurteil

Kaum ein anderes technologisches Thema heutzutage wird hinsichtlich gesellschaftlicher Auswirkungen so kontrovers diskutiert wie das der Künstlichen Intelligenz (KI). Während das Wörtchen „KI“ bei den einen Zukunftsvisionen hervorruft, in welchen technologischer Fortschritt menschliche Probleme wie Hunger, Krankheit und Klimawandel reduziert hat, zeichnen andere düstere Bilder von Orwell‘schen Überwachungsstaaten und technologischen Apokalypsen.

Starke, schwache KI

Es ist die Unschärfe des Begriffes „KI“, welcher eine derart große Bandbreite an Zukunftsszenarien ermöglicht. Für diejenigen, welche sich an solch spekulativen Debatten beteiligen, beutet KI „starke KI“ – eine künstliche Intelligenz, deren intellektuellen Fähigkeiten die eines Menschen erreichen oder gar übertreffen. Und so spannend die Diskussion über starke KI auch ist – sie ist reine Spekulation. Heute existierende KI ist weit, sehr weit von starker KI entfernt. Worüber wir heutzutage verfügen ist die sogenannte „schwache KI“ – Algorithmen, die spezifische Anwendungsprobleme (z.B. Bilderkennung, Spracherkennung, Übersetzung, Go spielen) lösen können. Und das mitunter sehr viel besser als Menschen.

Wo heutzutage „KI“ draufsteht, sind innen überwiegend Algorithmen aus dem Bereich des maschinellen Lernens (allen voran Deep Learning) am Werk. Diese Algorithmen können selbständig die Vorgehensweise erlernen, die zum Beispiel nötig ist, um einen gegebenen Input (z.B. ein Bild) auf einen gegebenen Output (z.B. eine Kategorie, welche den Bildinhalt beschreibt) abzubilden. Aber selbst diese „schwache KI“ birgt beträchtliches Potential – denken wir an mögliche Verbesserungen z.B. im Bereich der Medizin, Logistik, Verkehrssicherheit oder Energie- und Ressourcennutzung! Angesichts der Chancen, heutige Prozesse und Anwendungen zu verbessern, haben wir allen Grund, dem Einsatz von KI aufgeschlossen gegenüber zu stehen. Vorausgesetzt natürlich, dass KI verantwortungsvoll, „ethisch“ und sicher eingesetzt wird.

KI auf Abwegen

Ethische Herausforderungen von KI ergeben sich dabei zum einen durch die Zielsetzung. Wie ein Hammer für den Nagel an der Wand oder für den Hinterkopf eines Gegners verwendet werden kann, kann auch KI für böse Ziele missbraucht werden. Nur, dass KI im Zweifel deutlich größeren Schaden anrichten kann als ein einfacher Hammer. Und so sollten wir angesichts der Risiken dringend international diskutieren, wie wir uns hinsichtlich militärischer Anwendungen von KI verhalten wollen.

Zum anderen dringen besonders aus den USA, in denen KI Algorithmen schon heute in deutlich größerem Ausmaß eingesetzt werden als in Deutschland, immer wieder beunruhigende Nachrichten über voreingenommene KI Algorithmen. Zum ersten fand eine Studie kürzlich heraus, dass kommerziell erhältliche Gesichtserkennungsalgorithmen für Frauen bzw. dunkelhäutige Menschen schlechter funktionieren als für Männer bzw. hellhäutige Menschen. Mit der unschönen Konsequenz, dass es z.B. bei einem Abgleich mit Verbrecherfotos bei Menschen mit dunkler Hautfarbe deutlich häufiger zu falschen Übereinstimmungen kommen kann als bei Menschen heller Hautfarbe. Zum zweiten wurde vor kurzem bekannt, dass eine experimentell von einem großen Technologiekonzern zur Bewertung von Bewerbungen verwendete KI von Frauen stammende Bewerbungen systematisch schlechter bewertete als von Männern stammende Bewerbungen.

Wie KI zu Vorurteilen kommt

Um die Ursachen für vorurteilsbehaftete KI besser zu verstehen, lohnt es sich, einen Blick hinter die Kulissen zu werfen. Denn wie jede Technologie existiert auch KI nicht im luftleeren Raum. Dies lässt sich leicht anhand der Faktoren verdeutlichen, welche zum Erfolg heutiger KI beigetragen haben: bessere Hardware, cleverere Algorithmen und größere Datenmengen. Und gerade diese Daten sind es, durch welche Vorurteile in KI Einzug halten können.

Die Vorstellung von „neutralen Daten“ ist nämlich eine Wunschvorstellung. Im besten Fall spiegeln Daten die Welt wider, in der wir leben.       Eine Welt zum Beispiel, in der in Technologiekonzernen typischerweise deutlich mehr Männer beschäftigt sind als Frauen – was eine auf dem Personalbestand eines Technologiekonzerns trainierte KI dazu veranlassen kann, zu „schlussfolgern“, dass männliche Bewerber im Auswahlverfahren zu bevorzugen sind. Oder eine Welt, in der Länder bzw. gesellschaftliche Schichten innerhalb eines Landes unterschiedlichen Zugang zu modernen Technologien oder auch Bildung haben. Eine Ungleichheit, die sich als Dominanz westlicher Industrienationen in der geografischen Zusammensetzung von zum Training von KI-Algorithmen verwendeter Datensätze auswirken kann. Eine Dominanz, die wiederum zur Folge haben kann, dass derart trainierte KI-Algorithmen besonders gut für Menschen aus westlichen Industrienationen funktionieren. Ganz zu schweigen von der Voreingenommenheit der menschlichen Wahrnehmung, welche die Zusammensetzung von Daten beeinflusst – denken wir an das begrenzte Spektrum der Bilder, welche uns zuerst zu dem Begriff „Genie“ in den Sinn kommen.

Aber nicht nur die verwendeten Trainingsdaten, sondern auch bei der Entwicklung von KI getroffenen Design-Entscheidungen können negative Auswirkungen haben. Wenn bei einem nicht perfekt funktionierenden Bilderkennungsalgorithmus potentiell abwertende Kategorien zur Klassifikation zur Verfügung stehen, kann dies dazu führen, dass – wie in der Vergangenheit geschehen – dunkelhäutige Menschen als Gorillas klassifiziert werden. Wenn bei der Evaluation eines z.B. für die Gesichtserkennung eingesetzten KI-Algorithmus nur die Genauigkeit über alle Bevölkerungsgruppen hinweg berücksichtigt wird, können Ungleichheiten in der Genauigkeit nicht entdeckt werden, was zu Problemen bei der Anwendung führen kann. Denn Nutzer von KI-Algorithmen vermuten zumeist, dass die Algorithmen für alle denkbaren Anwendungszwecke geeignet sind.

Werte statt Wegsehen

Entgegen der verbreiteten Auffassung sind KI Algorithmen also nicht notwendigerweise vorurteilsfrei – sie können menschliche Voreingenommenheit bzw. gesellschaftliche Ungleichheit widerspiegeln. Da Algorithmen anders und in anderem Maß als Menschen eingesetzt werden, kann das bei blauäugiger Verwendung dazu führen, dass bestehende Ungleichheiten nicht nur bestärkt, sondern sogar vergrößert werden. Richtig angewendet können Algorithmen jedoch helfen, implizite und explizite Vorurteile menschlicher Entscheider zu mindern. Denn wie wir durch viele Studien wissen, ist die Liste der kognitiven Verzerrungen, die wir Menschen aufweisen, lang.

Es ist für den verantwortlichen Einsatz von KI in einem sensiblen Kontext somit essenziell, zu wissen, welche „ethischen“ Kriterien KI für den konkreten Anwendungsfall erfüllen muss. So kann sichergestellt werden, dass die KI den Anforderungen entspricht, bevor sie angewendet wird – oder aber, dass sie solange nicht angewendet wird, wie sie den Anforderungen nicht entspricht. Und mittels Transparenz, Überwachung und Feedback-Möglichkeiten lässt sich vermeiden, dass ein selbst-verbessernder KI-Algorithmus im Laufe der Zeit das ihm gesteckte Ziel verfehlt.

Für viele Anwendungsfälle sind derartige ethische Fragen jedoch vernachlässigbar, denken wir zum Beispiel an die Vorhersage von Maschinenausfällen oder die Extraktion strukturierter Daten aus unstrukturierten Dokumenten. Aber es ist nichtsdestotrotz gut und wichtig, Ethik und KI zusammen zu denken. Denn dies ermöglicht es uns, sicherzustellen, dass wir KI auf die bestmögliche Weise einsetzen. Denn das enorme Potential von KI gibt uns die Chance, den Status quo nachhaltig positiv zu verändern – technologisch wie ethisch.

WordPress › Error

There has been a critical error on your website.

Learn more about debugging in WordPress.