Algorithmic bias in artificial-intelligence (AI) and machine-learning (ML) systems has come under scrutiny over the last couple of years. While much of the focus is on the negative consequences of bias, marketers can also use bias in positive ways. But to do so, the bias needs to be recognised and understood, which starts with understanding how it originates.
AI bias essentially means AI or ML is making decisions with a certain bias towards a specific outcome or relying on a subset of features. A common example is a facial-recognition system that has been trained with mainly Caucasian people. As a result, the system cannot make accurate judgements about people from different cultural groups.
The model is making decisions from a set of features that is not representative of the data it is meant to be making decisions with, or it performs badly on some types of data the model has not seen during training.
Where does AI bias come from?
AI and ML systems are trained using sets of data that are acquired using various mechanisms. The data that is used to train the system inputs or features are used to make decisions. These outputs are sometimes called labels. Some sets of features can be biased towards specific results. In these cases, the systems perform poorly on some types of data they haven't been exposed to during training and give results that are not optimised.
It is important to note that the ML model itself is not the origin of the bias. The bias comes from the data that the model is trained with. In some cases, the system will perform very well on certain types of data but badly on others.
There are some clear examples of how bias in AI models can result in negative consequences. This has been seen in the United States where models trained with historical data that African American, Hispanic and other minorities have been overrepresented in crime statistics. This has resulted in AI models being used for sentences that lead to harsher penalties for members of those minorities.
If the data is skewed in one particular way, the model will make decisions based on that skewing of the data.
Not all bias is bad
However, for marketers, some bias in the data used to train AI models is valuable.
Bias is often discussed in a negative way, but people may want models to push in a particular direction rather than be absolutely neutral. If everything is neutral, then models face much harder learning tasks.
This is because an absolutely neutral model can take a very long time to curate the right set of training data in order to deliver benefits. If you are serving a particular customer base, then training the AI model with data for that customer base can be highly beneficial. Exploiting bias can help ensure your AI model delivers value from the outset of its deployment.
For example, if a business is selling fashion products that are targeted at young women aged between 18 and 25, then a recommendation engine that is powered by AI can use inherent bias to suggest further purchases based on other people in that target customer group. As the customer makes more decisions, the model can learn the preferences of that customer and deliver better targeted suggestions.
To get stronger performance at the beginning of the model’s use, some bias is useful because it helps to maximize return from the model initially. When the model used is serving data that has the same bias as the training data, this bias can be exploited so that the model performance is good from the start.
Leveraging AI bias in marketing
Once the model has been in operation, marketers can make more accurate decisions by using data collected as the model has been working. For example, a recommendation engine may start by making suggestions based on people the model believes are like your customers. But then, as the model learns more about your actual customers, it can make recommendations that are more specific to them.
By using the bias in the data, it is possible to reduce the initial costs of deploying the AI, as the cost of collecting unbiased data is higher.
For example, if marketers want to sell ads for cosmetics, then they should exploit selling to women and girls in the beginning. Then, to keep increasing volume, they can figure out what additional features they need to suggest cosmetics to men.
While exploiting the bias in the data used to train AI and ML models can be beneficial, it is important to recognise that bias in data can also lead to negative consequences. If bias is exploited at the beginning and then marketers lock into that bias, they might not see further improvements. For example, if you exploit a certain age group and do very well, after a while you realise your volume cannot increase anymore.
If you don't do something to overcome this bias, the campaign will gradually become harder to scale and more costly due to competition, since you may think only this certain group has the best performance, and therefore you only target them.
Overcoming AI bias
When AI and ML systems are trained using biased data and that bias is not recognised and addressed, there can be significant consequences. You may miss a potentially valuable customer niche and fail to continuously scale your market share. Being able to assess this and take action is critical.
One way to do this is to change the way that data is collected and see if this has an impact on model performance. Marketers can then conduct an A/B test, testing the model with different data sets to see which delivers better outcomes. As well as offering a path to optimization, this ensures new data does not decrease the effectiveness of the model.
Although refining the data collection is critical, it can be very costly without more insight. It’s important to evaluate how the model values specific features and combinations of features. By using domain knowledge, the model can be further refined. Deciding whether to refine the model or refine the data collection is a decision based on return on investment, where the expense of either changing the data collection methodology or reevaluating the importance of features in the model needs to be weighed.
It is important to understand that once a ML model starts operating, its initial results will be directed by the data that it is trained with. However, once the model is in operation, the system itself can continue to collect data and learn.
We certainly see this with online advertising. The machine, using the data it is trained with, will determine where to place a particular advertisement. Based on how users interact, the model will learn where to place ads in the future.
AI and ML algorithm bias is a challenge, but marketers who are aware of the implications of bias can be prepared and use it as a tool. When bias is understood, it can be used to assist AI models in their initial operating phase to deliver recommendations before the model learns from more data it collects in the field. But when it is not recognized, it can lead to unwanted results.
Dr Min Sun is chief AI scientist at Appier.