Popular AI models like machine and deep learning often result in a “black box†situation from their algorithms’ use of inference rather than actual knowledge to identify patterns and leverage information. Marco Varone, Founder & CTO, Expert.ai, shares how a hybrid approach using symbolic AI can help.
As AI becomes more integrated into enterprises, a substantially unknown aspect of the technology is emerging – it is difficult, if not impossible, for knowledge workers (or anybody else) to understand why it behaves the way it does.
This can create serious negative consequences for the operational models that AI influences because you can’t control a technology solution if you don’t know how it works. In turn, this diminishes the trust that AI needs to be effective for users. Let’s not forget that this particular technology already has to work with a substantial trust deficit given the debate around bias in data sets and algorithms, let alone the joke about its capacity to supplant humankind as the ruler of the planet. This mistrust leads to operational risks that can devalue the entire business model.
Black Box AI
The source of this mistrust lies in the algorithms used in the most common AI models like machine learning (ML) and deep learning (DL). These are often described as the “black box†of AI because their models are usually trained to use inference rather than actual knowledge to identify patterns and leverage information. In addition to this, by design, most models must be rebuilt from scratch whenever they produce inaccurate or undesirable results, which only increases costs and breeds frustration that can hamper AI’s adoption in the knowledge workforce.
This is why many forward-leaning companies are scaling back on single-model AI deployments in favor of a hybrid approach, particularly for the most complex problem that AI tries to address – natural language understanding (NLU).Â
The only way to solve real language understanding problems, which enterprises need to tackle to obtain measurable ROI on their AI investments, is to combine symbolic AI with other techniques based on ML to get the best of both worlds. Being the first technology created and widely used to mimic human understanding of language, it is not a limitation but a significant value addition because it is well-known and can be used in predictable and explainable ways (no “black boxes†here). It uses explicit knowledge to understand language and still has plenty of space for significant evolution.
Now that AI is increasingly being called upon to interact with humans, a more logical, knowledge-based approach is needed.
See More: What Is Artificial Intelligence (AI) as a Service? Definition, Architecture, and Trends
Following the Rules
Symbolic AI is built around a rule-based model that enables greater visibility into its operations and decision-making processes. In this way, operators can quickly analyze their operational patterns to detect errors and other anomalies in the data and the algorithm itself. It is important to note that, these days, rules can be generated automatically (based on ML techniques) starting from a set of annotated content, with the same process of ML only approach but obtaining a “white box†that can be understood and modified at any single level.Â
Thus, standard learning algorithms are improved by fostering a greater understanding of what happens between input and output. In the black box world of ML and DL, changes to input data can cause models to drift, but without a deep analysis of the system, it is impossible to determine the root cause of these changes.Â
Lines of Communication
At the same time, symbolic AI allows for better communication between humans and computers. Enterprises in industries as diverse as insurance, finance and media have deployed all forms of AI to better interpret the growing volumes of unstructured data found in documents, emails, social media content and similar communications.
By bridging the divide between spoken or written communication and the digital language of computers, we gain greater insight into what is happening within intelligent technologies – even as those technologies gain a firmer grasp of what humans are saying and doing.Â
See More: What Is Deep Learning? Definition, Techniques, and Use Cases
The Bottom Line
For the enterprise, the bottom line for AI is how well it improves the business model. While there are many success stories detailing the way AI has helped automate processes, streamline workflows and otherwise boost productivity and profitability, the fact is that a vast majority of AI projects fail. In case of a failure, managers invest substantial amounts of time and money breaking the models down and running deep-dive analytics to see exactly what went wrong.
With a hybrid approach featuring symbolic AI, the cost of AI goes down while the efficacy goes up, and even when it fails, there is a ready means to learn from that failure and turn it into success quickly.
Do you think the cost and efficacy benefits of symbolic AI are significant enough to warrant a shift? Tell us what you think on LinkedInOpens a new window , TwitterOpens a new window , or FacebookOpens a new window . We love it when you share with us!