Every time I mention Quantum Computing, I get a reaction. Sometimes its "Oh, you must be really smart to understand it", sometimes it is "Fascinating! How can I learn more?" often it is "Sure that sounds glorious but how does it affect my daily life?" - fair. With every major tech company
making wave after wave after wave lately, its natural to feel like its a passing tech hype, joining the likes of Blackberry and Bitcoin.
Its times like this that it is wise to look back. Can you imagine living on this planet a century ago? Yeah, me neither. There were neither computers, nor cellphones and electricity, light bulbs and air planes were the bleeding edge inventions; the thought of instant communication across the globe was unfathomable. Fast forward to today where we take these things for granted, thanks to innovators like Alan Turing, Von Neumann, Gordon Moore, William Shockley, Grace Hopper among many others. Quantum computing seems far fetched at the moment, just like how a century ago people were as pessimistic about the Digital age. If you want to be on the right side of history, you should care about quantum computing.
Is quantum computing any better than classical computing? What makes it stand out? In a nutshell, classical computers encodes information as bits.
A bit can be in one of two states: 0 (or off) or 1 (or on). Every piece of data and instruction can then be mapped to a series of bits that the computer can understand and manipulate. Computers understand electricity; presence of it maps to 1 state and lack of it maps to 0 state. Computers can manipulate bits using logic gates that perform Boolean operations such as AND, OR and NOT. These gates are made possible by the transistor, probably the most important invention in modern history.
In 1965, Intel CEO Gordon Moore predicted that the performance of computers would double every 2 years and volumes would decrease. This prediction has been holding true for about 50 years now as transistors are made more powerful and tinier. More transistors can be packed into Integrated Circuits (IC's) but we are now reaching physical limits, further compression of chips leads to electrical leakage and overheating. Moreover more investment leads to diminishing returns as the longevity of computers increases.The next major leap in computing requires a paradigm shift.
To understand quantum computing you have to forget all (or most) you've ever known about classical computing. A Quantum bit, also known as a Qubit, can be in states 1, 0 and somewhere in between (also known as superposition). A good analogy is a coin which can be heads, tails or flipping between the faces. When a qubit is measured, it collapses to state 1 or 0. Since observing a qubit alters its state, we can use this property to detect whether information has been compromised. Another interesting quantum property is entanglement, the idea that two or more qubits can be highly correlated, irrespective of distance. Entangled qubits can be miles apart but any effect applied to one will affect all of them, sort of like identical twins. With superposition and entanglement, a quantum system can be in multiple states at the same time, allowing a quantum computer to perform multiple calculations simultaneously.
With the fundamental properties described, quantum computers are better suited for certain applications than classical ones. Here are just a few"
Web applications use public key cryptography to ensure secure transmission of data over networks. In the wild, packets have to encrypted so that if they are sniffed, the interceptor can't make sense of it. Remember in elementary school when you had a secret code with your friend? You could safely pass notes to each other across the classroom because you were sure no one would understand your code; same concept is needed for the web. Classroom code works because you and your friend know each other and agreed on the code before hand. The web on the other hand, is made of strangers, you can't agree on a code with a retail website in person, the secret code, also known as private key, has to be transmitted over the same web. If an attacker intercepts both the key and message, your data is compromised (think credit cards, passwords etc). Since it is very hard to securely transmit private keys, public key cryptography is widely used instead. In this model, servers advertise public keys that anyone can use to encrypt messages to them. They then de-crypt these messages using a private key. The public key is product of two large prime numbers, from which the private key can be deduced. The only reason public keys are secure is because factoring them is very hard on classical computers. On the other hand, factoring is trivial using Shor's algorithm on a quantum computer; a hacker with a quantum computer is a deadly weapon. Quantum computing poses this problem and offers a solution too. Entangled qubits are highly correlerated and measuring a qubit collapses (alters) its state. Private keys can be transmitted with entangled qubits; if an attacker gets to observe any of them, their states will be compromised leading to detection. Evolution of quantum computers will definitely raise the bar for cryptography.
It is possible to simulate quantum systems on classical computers, but it is not the most efficient. It takes exponential time in number of particles to perform simulations on classical computers. In 1982 Richard Feynman and Yuri Manin suggested building quantum simulators to speed up the process. These simulators are different from generally programmable quantum computers but they are essential to understanding and solving physics problems such as many-body physics and low-temperature physics. Similarly quantum molecular simulation will advance drug discovery and development.
The ability of quantum computers to perform multiple operations simultaneously makes them well suited for optimization problems such as Travelling Salesman Problem and pattern recognition. These problems can be accelerated because solving them involves searching the problem space for the most optimal (minimal error) solution. There is a symbiotic relationship between quantum computing and machine learning. Machine learning helps us understand quantum systems better, quantum computers, along side classical computers, accelerate machine-learning computations. One could say I'm just stringing up buzzwords together, but it is happening. These are exciting times for the Information Age.
If quantum computers are so great why are they not commercially available? There is a number of quantum computers developed by research institutions such as IBM, D-Wave Systems and Rigetti among others. They are just not powerful enough to replace classical computers yet. Qubits are harder to manipulate since they are sensitive to disturbances, little disturbance can cause them to fall out of their state (decoherence). It is possible to mitigate this effect by quantum error-correcting and researchers are on the race to quantum computers a practical reality.
A century ago, computers were far from imagination. Alan Turing didn't envision a n era when cellphones could replace calendars, notepads and books yet here we are. Likewise, there is no telling how far quantum computing will take us in the next century, we have barely scratched the surface. Join the revolution!