Artificial Neural Networks and Leadership Not So Different

If you are a leader who would like to gain a deeper understanding of artificial neural networks by using your leadership experience rather than reading volumes of literature on the subject, if you are a data scientist looking to leverage your knowledge of artificial neural networks for a top-level leadership position, or if you are just curious about a fresh perspective on artificial neural networks, this post is for you.

In case you don't see how artificial neural networks and leadership are linked, let's dive right in with a #TechGrowthMindSet hat.

#TechGrowthMindSet is an interdisciplinary approach I developed that teaches us new things by drawing parallels with other areas of our professional and personal lives. Using this method, we not only ramp up quickly with the unfamiliar, learn new technologies based on different aspects of our day-to-day activities, but we also take some of the lessons learned and implement them back in our everyday lives. For example, by observing the benefits of creating resilient technical solutions, we may also be able to appreciate and resolve to improve our own resilience. I've written about my experiences with the #TechGrowthMindSet in some of my previous blog posts on topics such as Blockchain, Cloud Architectures, Metrics, etc. Many of the lessons I learned could be applied to self-improvement.

I lead several initiatives at Cisco's Emerging Technology and Incubation team that leverage AI/ML and neural networks to deliver robust, transformational solutions to customers, propelling Cisco's business into new areas. To understand this technology quickly, I asked myself, "What do I already know that can be applied to ANNs. How can it help me better understand what, why, and how ANNs function?". In this post, I'd like to share what I've discovered.

Connecting The Dots- Artificial Neural Networks & Leadership

It turns out that architecting and deploying an artificial neural network is very similar to practicing a successful leadership strategy. Foundational principles such as adaptive learning through feedback make them very similar. To improve business outcomes, successful leaders actively solicit feedback from their customers and teams. Neural networks depend on feedback loops to deliver valuable insights. A good neural network has to be efficient, accurate, verifiable, robust, ethical, tamper-proof, explainable, just as a good leadership approach has to be nimble, scalable, effective yet maintain transparency and integrity. As good leaders can thrive within the messiness, navigating ambiguity, and rapidly leading change within organizations to deliver results, artificial neural networks outperform other machine learning techniques when dealing with messy volumes of unstructured data such as images, audio, and text.

You can skip the intro section and go straight to the next section if you know the basics of neural networks.

Introduction To Artificial Neural Networks

As breakthroughs in artificial neural networks have made way for deep learning in recent years, companies across all industries are implementing this technology as part of their AI strategy. Deep learning has enabled a wide range of new advanced smart services in consumer and industrial applications from sentiment analysis in customer service to image recognition in retail, surveillance, autonomous vehicle driver safety, healthcare, and so on.

Artificial neural networks are a powerful AI approach that allows computers to learn from input data without having to be explicitly programmed. The neural network draws inspiration from the biology of the brain, can learn on its own, making machines more human-like, transmitting information between layers of so-called artificial neurons, hence better known as Artificial Neural Network (ANN).

ANNs take a layered approach to process information and make decisions. In its basic form, an ANN can have only three layers of neurons: the input layer (data ingestion point), the hidden layer (processing point), and the output layer (decision point). ANNs can often get more complex with multiple hidden layers. When there are more than three layers, we refer to it as deep learning. Whether it’s three layers or more, information flows from one layer to another creating a relay of information flow. Deep artificial neural networks are a set of algorithms that have set new records in accuracy.

For a neural network to learn, there must be an element of feedback involved—in a similar way as humans learn by being told what they are doing right or wrong. Once you have trained the artificial neural network with enough learning examples, it reaches a point where you can present it with an entirely new set of inputs it has never seen before and examine how it responds.

Artificial Neural Networks: A Leadership Approach: 12 Steps

To further crystallize the idea, let’s break up ANN components at various phases of the development process and see how they relate to a leadership strategy in a simple step-by-step approach. Each action of an ANN in the figure below corresponds to a leadership action shown on the right. This process on the left is generally applicable to designing any AI/ML models not limited to neural networks. Feedback loops are predominantly characteristic of ANNs.

We will explore these steps in more detail in future posts. I hope the new perspective of viewing ANNs from a leadership standpoint was helpful. Because #TechGrowthMindSet is bidirectional, leaders can also learn from ANNs and their efficiency. After learning about ANNs, it would be interesting to see if any leaders consider assessing their leadership approach to improve current practices based on these insights.

In our team, we utilize heavy loads of AI, machine learning, and deep learning both at the edge and in the cloud. A mix of AI/ML experts and novices work together on innovation projects to make breakthrough deliverables possible. Leave a comment about how you have been handling rapid learning and the application of AI and ANNs to solve your organization's business-critical problems. What methods have you used to become familiar with this technology?