Scientific Computing: Artificial Neural Network

Artificial Neural Networks are computational models that are capable of machine learning and pattern recognition.  They were inspired by the animal central nervous system. Today, they are used in many fields and have been well-established in any discipline, especially medicine.  They can extract information from raw data and then build computer models for medical decision-making.
Now you know what they are capable of doing but what exactly are they? They  are basically systems of connected neurons that compute values from inputs by pushing information through the network.  According to the Wiley Online Library, these models contain computing nodes that operate as nonlinear summing devices.  The nodes are interconnected by weighted connection lines, and the weights are adjusted when data are presented to the network.  If successfully trained, the artificial neural networks can perform tasks such as classifying an object, predicting an outcome value, recognizing a pattern, and completing a known pattern.  

Since applications of artificial neural networks serve such broad spectrum of fields, they help validate research in those areas.  For example, medical decision support in medical applications.  These networks must be understood thoroughly before being used. Check out this list in which neural networks have been successfully applied.  

Computer Graphics: Computer-generated Imagery

Have you seen the latest Ender's game movie? No!? What are you doing reading this post then? In most movies like Ender's game, computer-generated imagery (CGI) is used.  It is an application of Computer Science (commonly referred to 3D graphics) to create special effects in films, television shows, simulators, commercials, etc.  Content is able to be produced without the use of extra props, and expensive set pieces. The real question is how are these special effects done? There are a few steps to incorporate CGI into a film:

1. Modeling
2. Animation
3. Rendering
4. Compositing

The first step is for an artist to create the images the director wants by the use of clay models, photographs, or drawings.  This could be the hybrid human-alien race in the movie Avatar in which actors take the place of them.  

The second step starts when the model is perfected. The animator tells the computer how the model moves.  An example of this could be how the Transformers move or how their face moves when they talk.
The next step is for a renderer (program that contains lighting, texture, shading, and more features) to create images showing the moving 3D models.  The director modifies the lighting and materials, often doing so until it looks perfect. 

Lastly, the CG images are combined with the live-action part using compositing software.

Communications and Security: Computer Security

Our idea on computer security has changed a lot in the recent years.  When people thought of computer security, they thought about protecting the physical machine that contained important data.  However, nowaday only the hardware is secure and the data is vulnerable to attacks.  With each major technological advance in computing comes more security threats. Technology advances too fast for solutions to be developed.  Therefore, developing security is just too difficult.  The penetrator tries to find the holes while the designer tries to close them.  The issue here is that we can never know if a system is perfectly secure, and that there is a likelihood that the penetrator will find a hole.  Of course, with the appropriate techniques, the system can be built so it is highly unlikely to penetrate its security.  BUT who prioritizes securitiy? When looking at a system, people often look at its price, functions, performance, reliability, compatibility, etc.  Security will be at the bottom of the list!  It may even just be a nuisance.

The bottom line is to protect yourself.  The problem is not the computer, but THE PEOPLE. Download anti-virus programs to prevent your personal information from being stolen.

Artificial Intelligence: Video Games

Can you think of a game without an AI or computer controlled opponent? I certainly can't! One of the most interesting fields of Computer Science is Artificial Intelligence, or known as AI.  There are many applications for AI such as speech recognition (like Siri on iPhone), natural language processing, computer vision, etc.  However, I find games are some of the more interesting applications that utilize AI.  Well what is AI in games? They are techniques used to produce the illusion of intelligence in the behavior of non-player characters.  You could have competent opponents or helpful allies!

How it is implemented depends on the type of game, such as RTS (Real-time strategy), FPS (First-Person shooter), RPGs, sport games, etc.  Here are some common elements in which AI are implemented:

• Event Driven Engine: AI reacts based on events.
• Specific AI Structure: how the AI moves and determines reasoning and its behavior.
• Path-Finding: finding the shortest path to a certain destination.
• Animation sequences for different body parts such as shooting/reloading in FPS.
• Scripted Behavior: show their personality through dialogue, contributing to the overall storyline of the game.
• Collision: detect obstacles in game.

As an SJSU student, I have yet to learn about the algorithms used in video games to create AI. The most popular algorithms used are A(*),  Finite State machines, and Artificial Neural Networks.  Every game you've played has definitely used some sort of AI, and have implemented these algorithms.  

History of Computer Science: Mobile Application Development

How long ago do you think Apple released the Itunes App Store? Only five years, but it feels like ten years to me! In just five years, smartphones have become ubiquitous as well as its mobile applications. These mobile applications have now become a great part of our daily lives. The birth of the Itunes App Store in 2008 led to the creation of the millions of apps that we have today.  What apps you may ask? Well, a well-designed app has the potential to help increase productivity and even revolutionize the job market.

The history of mobile application development actually dates back to 2003, introduced first by IBM. It was not popular back then and only had basic features.  However, a later major advancement arrived in 2002, when the Blackberry-smartphone was introduced.  Thus began wireless emailing.  As time went on, new mobile platforms(iOS and android) were introduced which further boosted mobile app development.  With these smartphones, mobile technology became completely revolutionized.  Mobile apps have allowed people to become more efficient and allow new experiences.  For me, I have used the Udemy app which allows me to take a course anywhere, without the need for laptop or desktop.

File Sharing: Client-Server Downloading vs. P2P Sharing

I’m sure you have downloaded something off the web but do you know the process?  When you download something, your computer, the client, tells the server (computer that holds the web page and file) to transfer a copy of the file to your computer.  This transfer goes through a protocol, or set of rules. The transfer speed is affected by the type of protocol (FTP or HTTP), amount of traffic on server, and number of other users downloading the file.  How does this traditional client-server download contrast with peer-to-peer sharing?

In peer-to-peer sharing, you use a software program, such as BitTorrent.  BitTorrent downloads torrents, which are files that point to the computer with the desired file. BitTorrent sends a request for the file by communicating with a tracker (central server with the file) to locate other computers connected to the internet.  The tracker finds a swarm, or a group of connected computers that are file sharing.  Once BitTorrent finds a seeder, or a computer that has the complete file, it will start downloading the file.   Basically, you are downloading small bits of files from different sources at the same time.   Once you finish downloading, you will become a seeder and help others download the file. 

BitTorrent is sometimes used for distributing copyrighted material but the program is still legal.  If you choose to download any copyrighted material, you are at risk of being caught, since the download can be traced back to you.  Be careful and happy downloading!

Data Structures: Real World Applications

If you’re a CS major, I am sure you have come across a data structure.  You may have worked with arrays, linked lists, graphs, or trees.  These are all ways of storing and organizing data for efficient usage. There is no “best” data structure and are used depending on the task at hand; all data structures have pros/cons.  The question is HOW are these structures used in the real world? Let’s take a look at a few common ones:

1) Linked List

An example on how a linked list may be used in the real world is managing a train.  These trains are linked together in a calculated way so that items can be loaded, transferred, unloaded, and picked up efficiently.  Like a linked list, trains can be added or removed at any point of the train.  Also, the train at the end can be removed easier than the train in the middle.  Therefore, one may use a linked list in order to plan the order of the trains, and the contents of the trains.  

2) Binary Search Trees

Binary Search Trees are often used for data that is constantly added or removed, such as a map or set of objects.  For example, any language library (C++, Java, .Net, etc) uses binary search to find items; items are inserted in order, and items can be easily accessed.  

3) Hash Tables 

Internet Routers use Hash Tables; a router table may contain millions of entries.  Therefore, the router must determine the best route by querying the router table when the packet is routed to a specific IP address.  In this case, the key is the IP address and the value is the address.

Knowing when to implement these data structures is important because some are more efficient, depending on the task!