Image Credit: KD Nuggets
When the majority of people think of artificial intelligence (AI), the images that come to mind are the autonomous beings in Star Wars or the virtual assistants from superhero movies. While we may not yet live in a future where AI has evolved to include semi-sentient beings, there are numerous advancements being made in AI that can greatly influence your daily life. Machine learning, for instance, involves the training of a program based on past data so that it can produce a model that can be used for later analysis.
There are a plethora of subsets of machine learning, but one that stands out from the others is reinforcement learning. Reinforcement learning algorithms are unique in their training of a model since they use an underlying process known as a reward system; as a model exhibits a series of actions in response to a certain task, it is either positively or negatively affected for its actions. This reward system ensures that, through successive iterations, the model is trained to maximize the number of rewards it receives.
Take AlphaGo, which is notable for its successes in the complex game of Go. By implementing reinforcement learning, the program instead aims to maximize the rewards it receives. However, the unit of this reward may vary per scenario; as stated by Martin Heller of InfoWorld, “AlphaGo maximizes the estimated probability of an eventual win to determine its next move. It doesn’t care whether it wins by one stone or 50 stones.” While the “reward” may sometimes be numerical, it can also be the likelihood of a favorable outcome, as illustrated here.
The underlying theory of reinforcement learning, however, emphasizes that the model is attempting to maximize the reward. Through the manipulation of several parameters and designing a model that responds to certain behaviors by increasing or decreasing a reward by a factor, researchers can ensure that the model mimics the characteristics that they are aiming for.
On a larger scale, one may take a look at Amazon, which is revolutionizing its delivery system through the use of drones that utilize reinforcement learning. According to Ron Schmelzer, Amazon “used machine learning to iterate and simulate over 50,000 configurations of drone design before choosing the optimal approach.” This particular use of reinforcement learning illustrates the efficiency through which designs and models can be adapted to a situation. Rather than a manual or basic algorithmic approach, Amazon’s use of a reward system in place of traditional methods emphasizes the possible future of reinforcement learning.
One of the most fascinating aspects of reinforcement learning is that there is a multitude of possibilities to explore and queries to test. While there are a plethora of uses for reinforcement learning currently, there are sure to be more in the future. So while we may not yet live in a society where communicating with robots is as simple as talking to other humans, we are definitely on our way to making significant advances in machine learning and artificial intelligence as a whole.
Hey Allgirlithm Readers! Hope you're as excited for summer as we are. (And for those still in school, keep pushing!) We're here with some cool opportunities and resources to help you keep up with your computer science studies. P.S., if you haven't checked out our AI Club program, you're missing out! Summer is a great time to start planning for projects during the school year, and we'd love to have you as part of the team. We're also rolling out a similar program for tech workshops and other outreach events, using open-sourced curriculum free and open to everyone, so stay tuned.
For a refresher on all things tech, check out this great resource again: https://code.likeagirl.io/a-high-school-students-guide-to-cs-programs-internships-487586031e07. You'll see a bunch of summer programs; although some deadlines have passed, the internship advice is also really great.
This is also a great time to look for scholarships. Check out this spreadsheet for a comprehensive list:
Need other opportunities? Look at NCWIT: https://www.aspirations.org/participate/opportunities
Or just want to get mentorship? Join #BuiltByGirls WAVE program:
For those in the Bay Area: Check out Bay Area Teen Science at http://bayareateenscience.org/, or give them a follow on social. They have some great opportunities specific to California.
The Congressional App Challenge is coming up in November. You will need submit an app, a demo video, and written responses to the competition, so it's great to start in the summer. Free resources include MIT AppInventor (http://appinventor.mit.edu/explore/) and Xcode (free download on Mac).
The NCWIT Aspirations in Computing Award application is due in the fall of 2019. The application includes several questions and essays. The award is based off of aspirations, so don’t worry if you are relatively new to coding!
That's all for now! As always, look at Allgirlithm's Resources page if you're stuck, or reach out to us at firstname.lastname@example.org
Photo credit: brilliant.org
A classic problem in probability, and one that has prompted numerous discussions and the use of simulations, the Monty Hall problem encourages the use of critical thinking in determining whether new evidence can alter one’s course of judgement.
Competitive programming problems are generally divided into a number of categories, each requiring a different skill set, algorithm, or data structure to solve. One subdivision of programming problem topics includes problems which are best solved using what’s known as a “greedy approach”.
A greedy algorithm makes locally optimal decisions, which hopefully produce a globally optimal solution. Furthermore, a greedy algorithm does not revisit a previously-made decision. These characteristics make greedy algorithms easy-to-code, straightforward, and efficient / fast. However, coming up with a greedy solution to a problem typically involves more algorithmic thinking; the difficulty in implementing a greedy approach lies in proving that it will work.
Applications are open for scholarships to Apple's annual developer's conference! Submissions are due March 24, 2019. To apply, create an interactive Xcode Playground and answer several questions. Apple will likely select ~300 from thousands, based off of numbers from previous years, and will provide these developers with a free ticket to WWDC. Scholarship winners will have the opportunity to meet new people, learn new things, and explore Apple's upcoming technologies. Past winners have had the chance to meet Tim Cook and Michelle Obama. Don't delay; submit your project today!
Dear Allgirlithm Readers,
We hope 2018 has been great for you, and that 2019 will be even better.
We have some exciting new updates from Allgirlithm we'd like to share.
1 - We have a new domain! In case you haven't noticed, you can access our new blog at allgirlithm.org.
2 - AI Club - All kudos to Joanna, who's directed this program from scratch. We've now grown this club to twenty (!) locations, including three countries. Thanks to all of you for helping us promote it, and if any of you are interested in getting involved (either with starting a club or helping develop curriculum/direct/outreach) please reach out to Joanna.
3 - Impact - Since Allgirlithm started, up until this year, we've had over 20,000 page views and over 7,000 unique visitors. We hope to keep growing even more this coming year, with all your help!
Additionally, we're going to start an Allgirlithm magazine! If this sounds interesting to you/you'd like to contribute, please shoot me an email. It will most likely have articles, news, and cool design ;).
Finally, if you haven't followed our social media yet please go do so (and ofc we'll follow back!): @allgirlithm for Instagram, Facebook, and Twitter.
Cognitive science is the study of thought, learning, and organization. The field of cognitive science works to answer difficult questions about the nature of thought, intelligence, and other parts of our mental lives.
Competitive programming problems are generally divided into a number categories, each requiring a different skill set, algorithm, or data structure to solve. One subdivision of programming problem topics contains problems for which there exists no general technique or algorithm, i.e. no well-studied solution. These are known as ad-hoc problems.
Each ad-hoc problem is unique, and requires a specialized approach. Some competitive programmers consider ad-hoc problems to be the easiest type of problem. In reality, ad-hoc problems can be easy or hard! For beginners just starting to code competitively, it may be useful to practice lots of these problems.
Here is a good source of ad-hoc problems to practice with: codechef.com/tags/problems/ad-hoc. For each problem, first try to "pseudocode" (planning your solution on paper in non-code words) and solve independently. If you find yourself stuck on a problem for longer than several hours, it may be helpful to consult the solution in order to thoroughly understand the general approach, then code it yourself without checking the solution.
Photo Credit: Sportradar
From behind the scenes, technology is one of the main factors that makes sports so exciting. The power of machine learning and artificial intelligence can help expand the potential of professional athletes by recording and analyzing infinite details at a quicker speed than any sports scientist or analyst could. For example, cameras, videos, and sensors can be applied to the body or equipment and collect an enormous quantity of valuable information. On top of boosting an athlete’s craft, artificial intelligence also can improve the business side to the sport industry, such as a team’s ticket sales, sponsorships, and merchandising.
Photo Credit: Shutterstock
Back in November 2017, Facebook announced that it was using artificial intelligence to help
flag suicidal users. The tech was used to identify posts, videos and live streams that could be
indicative of suicidal thoughts. The algorithm also prioritises the order in which the
Facebook team reviews these posts. According to Facebook, these accelerated reports get
reported to local authorities twice as fast as unaccelerated ones – which could save many