Machine Learning & Neural Networks

Quantum Machine Learning is about how quantum computers and other quantum information processors can learn patterns in data that cannot be learned by classical machine learning algorithms to solve real world problems.

Artificial Neural Networks

Artificial neural networks are the heart of machine learning algorithms and artificial intelligence. Historically, the simplest implementation of an artificial neuron traces back to the classical Rosenblatt’s “perceptron”, but its long term practical applications may be hindered by the fast scaling up of computational complexity, especially relevant for the training of multilayered perceptron networks. Here we introduce a quantum information-based algorithm implementing the quantum computer version of a binary-valued perceptron, which shows exponential advantage in storage resources over alternative realizations. That is why our Quantum Algorithms are able to beat the markets consistently.

“We created our QuantumAlgo series of algorithms which leverages simulated data from various types of market conditions to select the best placement and execution strategies to maximize returns. It then uses reinforcement learning, a subset of machine learning, to assess the performance and utilizes neural networks to consistently learn and refine its performance.”

 

Dr David Moche
CEO of Qfinity Labs

Next Generation Algorithms

Automated trading algorithms have undergone a multitude of evolution over the past 20 years. Let’s look how these algorithms have evolved over the past decade

1st-Generation

The 1st-generation algorithms usually consisted of only buy or sell orders with simple parameters. It can execute simple buy or sell orders with predetermined rules usually used in combination with simple indicators. It cannot analyze the current situation to use the best rules as the rules are hard coded and not dynamic.

2nd-Generation

2nd-generation algorithms deployed strategies to break up large orders as to reduce potential direct market impact. For example, selling 500 million worth of euros versus the dollars in a short period of time could cause the price to sharply decline and potentially scare away buyers. However, slicing the large order into smaller orders reduces immediate market impact and help the seller’s order complete at a higher VWAP price which could lead to potentially higher profit margins.

3rd-Generation 

The 3rd gen trading algorithms are based on quantum computers, which utilizes artificial neural networks with machine learning. 3rd gen algorithms have proven to be highly successful at specific tasks like pattern recognition, image classification, and decision making. This allows the algorithms in real time to think and make the best decision in the constantly changing financial markets.

Using the analogy of teaching a robot how to walk, 1st gen rules-based technology would program the robot to lift one leg followed by another to move forward. 2nd gen algorithms would show the robot hundreds and thousands of videos that demonstrate how to walk and allow it to learn. 3rd gen algorithms will emphasize improved upon 2nd gen, which throws the robot into different environments and forces it to learn how to walk under these environments with might include different wind speed, different floor textures and with obstacles being thrown at it from all directions.

Did you know?

 

“AlphaGo, an algorithm developed by Alphabet Inc beat the best competitive human player in the strategy game “Go”.  The algorithm used machine learning combined with neural network capabilities to learn predict the player’s every move in order to beat him. Qfinity Lab’s “QuantumAlgo” series of trading algorithms uses exactly the same technology to consistently beat the market. It is expected that the performance can be further improved with the launch of QuantumAlgo 2.0 in 2021.”