How does A.I. really affect people ?
0 likes167 views
The document explores the transformative potential of artificial intelligence (AI) and deep learning in various sectors, highlighting their ability to make predictions based on massive datasets. Key applications include improving stock market investments, enhancing healthcare diagnostics, optimizing educational outcomes, and refining weather forecasting. As AI technology advances, its deployment is likely to revolutionize industries, making predictions more accurate and personalized.
How does A.I. really affect people ?
- 1. A.I. CAN SEE INTO THE FUTURE What we’re about to explore are the possibilities of deep learning which are already enabling machines to peek into the future. Driven by data the predictions come with greater confidence than ever before, and the scope of growth is quite akin to magic! Let’s take a brief look at the technology, how it actually works, and some of it’s novel uses to improve our society! What’s A.I.? The human brain is considered very unique - it’s extents & ability to learn, adapt, infer, ideate etc. seems to be an exclusive privilege to the human species. So much so, that we’re at the brink of replicating and imparting our abilities into our own creations - machines. A.I.(artificial intelligence), Machine Learning or Deep Learning, whatever you want to call it, is progressing swiftly in this direction - transcending the limits of just executing commands or following rules, machines are now gaining the ability to LEARN various abilities simply by using experiencies. Oh, because machines are un-emotional & devoid of existential friction (thus far), AND since they are limited only by our highly advance cloud servers for computational power, they’re greatly superior with their learnings than us biologically gifted homo sapiens. While machines are not 100% error free & have their own requirements for appropriate operation, when applied well, the width & depth of their understanding is far greater than that of a human brain despite only being cheap emulations of the same.
- 2. So, how can A.I. see into the future? Here’s a humble attempt to present machine learning in a nutshell: Machine learning is typically a program takes INPUTS (usually huge amounts of data), runs them through various layers of NEURONS (like the ones in our brains), each of which run simple functions to assess them, finds patterns in them (overall probability), and finally provide an output. This process is then repeated to "TRAIN" the machine for optimization, and because the machine doesn’t inherently know right from wrong, providing as much context (historic data) as possible, significantly improves the machine’s ability to be accurate with it’s learning. For example, most of us have figured out by now that, when the sky is getting dull, the air becomes breezy & the birds start scattering back to their trees, rain is likely- further, if it is the monsoon season, we would know our probability to be right is much higher since we’ve learnt to expect rains during those times. Now just imagine this done by a computer with terabytes of information about every cloud movement & temperature change, and you can expect some scarily accurate predictions about the weather! Interesting? Well, there are many resources of various degrees of complexity & technicality out there for you to get deeper into deep learning if you wish to go further down this rabbit hole. Then, how & where does this really help? Considering the amount data, we have accumulated and are actively collecting, the use cases of Machine Learning are far and wide, currently being deployed in many fields including healthcare, stock markets, weather forecasting to name just a few. In each case, it’s ability to see into the future is slightly different but fundamentally using the same principle of identifying hidden patterns in large volumes of data to predict a various event.
- 3. Stock Markets: To over-simplify, the main challenges in the stock market are to identify which of the companies are likely to grow (therefore invest in) and to identify when a company’s stock value is likely to drop (so you can liquidate and shift your investments) - ultimately to get the best return on your investment. But the factors that influence the price of a company’s stock are numerous and mapping them to past trends is a task that human calculations can only get so far. Enter deep learning A.I. Deep learning has recently been deployed to soak in this data to make investments based on finding hidden patterns and the results are apparently mind blowing! - 'EquBot' has surpassed the expected returns of $40m over A YEAR from $2.5m by reaching the target in just A WEEK!! No more keeping those fingers crossed, maybe? Similarly, the banking industry is starting to use AI to predict frauds & also optimize their data to make smarter decisions in various aspects of their businesses. Healthcare: The world of healthcare is full of data - huge volumes of readings across numerous specializations from research and field work alike. While most healthcare is backed up heavily by science, it is essentially (highly intelligent) guesswork based on symptoms & results from tests made by doctors who have studied enough material to notice these patterns. Doesn’t this sound a lot like how AI works? Well, it is…and the applications in this field, then, are both revolutionary and critical in the journey of human existence… From being able to make more accurate diagnosis for otherwise difficult conditions at the earliest of stages, to enabling wearable-tech to predict when a person is likely to need medical attention based on their real time metrics, A.I. is turning out to be a savior in this very important industry with its growing ability to anticipate many aspects of the healthcare both at an industry and personal level. (https://deepmind.com/applied/deepmind-health/working-nhs/health-research-tomorrow/)
- 4. Education: Education is another industry with huge volumes of data with patterns that couldn’t be identified too easily by human intelligence. AI is being employed to use data of grades of thousands of students across schools & over years to optimize curriculum, predict remedial steps for every individual & even improve the methods of teaching. AI can potentially predict the journey of a student & suggest an academic journey that would best cater to their interests and strengths. So, the next Picasso isn’t spending more time than needed in gym class & the next Usain Bolt isn’t forced to take up advanced physics.(https://www.forbes.com/sites/theyec/2017/12/27/how-ai-impacts-education/#26fb5fe8792e) Weather: This is probably the industry with the largest amount of data for the longest time and the use of analytics have been crucial in predicting weather trends for decades. Now, with the advent of A.I., these techniques have become significantly superior - government & private companies alike are using deep learning algorithms to better process the massive amounts of data collected by the thousands of weather satellites every minute to generate better predictions for much longer windows into the future. We are likely to see a day where we know exactly when and where a natural disaster is going to happen years in advance, thus allowing for zero causalities and minimal infrastructure damage. This kind of prediction is invaluable to many other aspects of city planning & anthropology. E-Commerce: Now let’s look at how this tech is disrupting the world of online shopping. The day’s of guessing which segment is likely to buy a product to shift marketing budgets are on their way out with AI-driven analytics being able to accurately predict probabilities-to-buy at an individual level! It’s the end of generalization and the advent of personalization - companies like Boxx.ai are able to make recommendations to each individual based on behavior, time of day, day of month, the weather & other factors in real time. This could mean many things - from recommending the perfectly related products for each purchase, to creating relevant deals for each customer, Deep Learning is key to reducing high marketing spends, high
- 5. operation costs, customer attrition, business leakages and improve inventory management, customer engagement, conversions, retention, loyalty, increase ROI & even garner cultural progress! But effectively, whether you look at this from a customer’s point of view or from a business point of view, the benefits are huge! Others: From managing transportation hassles in big cities, to making video games really learn from your every move, from helping farmers plan their crops based on weather predictions, to assisting the mentally ill plan for another episode, the applications of predictive A.I. are not just vast, but incredibly useful - people are beginning to use this tech to crunch data in every field possible. With more and more companies providing A.I. based tools at affordable prices, the field is more than likely to sky-rocket in the coming years, especially since it’s growth would effectively catalyze its own growth. Conclusion: AI & Deep Learning is an exciting step towards pushing the world to the next-level of progress - Machine Learning is probably going to be that critical step in this age of tech that will overhaul all other industries and technologies to become highly optimized. And whether it’s really looking into the future or just exercising the extents of computation on big data, the end result seems just as promising. Based on your high IQ, the humor woven into the prose, the good progress across multiple industries, our A.I. predicts that you are 83.9% likely to enjoy this article and therefore 97% likely to share this with your family & friends! Due credit to all the humorists featured here, all of whose other works are well worth checking out if you found any of these funny! Research references: https://www.teachthought.com/the-future-of-learning/10-roles-for-artificial-intelligence-in- education/ https://nypost.com/2017/12/06/ai-is-better-at-picking-stocks-than-the-stock-market/ https://www.pwc.com/gx/en/industries/healthcare/publications/ai-robotics-new- health/transforming-healthcare.html http://cs.stanford.edu/people/karpathy/convnetjs/demo/classify2d.html