Science Strikes Back: AI & Machine Learning
July 27, 2020 | Abree Murch
Our “Science Strikes Back” series introduces some of the most exciting, forward-leaning venture investment sectors in deep tech. Our Deep Tech Fund co-invests alongside established deep tech venture firms in companies taking on the toughest and potentially most lucrative technological challenges. Our point of view is that the need has never been higher for scientific thought and innovations tackling hard problems. The fund is now open! Click below to learn more.
Deep Tech Focus #5 – Artificial Intelligence & Machine Learning
Defining the Space
“The artificial intelligence problem is…making a machine behave in ways that would be called intelligent if a human were so behaving.”
– John McCarthy, coined the term AI
Artificial intelligence (AI) is machines doing tasks that usually demand human thinking or abilities, demonstrating intelligent problem-solving and learning, for example. Importantly, like humans, the machine has the ability to improve over time.
AI has been enabled by the tremendous increase in available data, advanced algorithms, and much greater computing power and storage. Yet, even given those advances, machines are still in the “narrow” category of ability — that is, they’re better than humans on one particular task. Still ahead: when machines are the equivalent of humans in any thinking tasks (“general”) and then superior in many tasks (“strong”). Examples of AI include the Roomba vacuum, Betterment robo-adviser, and Google maps.
Machine learning (ML) is one example of AI. It refers to the ability for computers to analyze data, classify it, recognize patterns, and make predictions. This “thinking” is performed without the explicit direction of any programming. In other words, machines learn for themselves without human intervention. Examples of machine learning: Netflix or Amazon recommendations, spam detection, chatbots, and self-driving vehicles.
Applications and Benefits
AI and machine learning have numerous applications across industries.
- Auto industry: Self-driving cars, co-pilots
- Entertainment and retail: Assessing user activity and recommending products
- Manufacturing: Supply chain monitoring, predicting market demand, monitoring and performance assessments
- Healthcare: Analyzing massive data sets and spotting patterns
- Finance: Analytics, autonomous trading and portfolio management, fraud detection
- Data security: Bug detection, cyberattack prediction
The benefits of this technology include:
- Superior predictive ability
- Increased speed, efficiency, proficiency
- Assessing more data in less time and identifying insights or patterns
- No fatigue
- Frees up humans from repetitive tasks
- Increased safety
Innovations in the Wings
- Natural Language Processing focused on automated understanding and generating data from speech and text will produce better customer understanding and service.
- Computer Vision will analyze and label images to improve comprehension of the visual world.
- Technology will be programmed with Empathetic AI.
- Edge AI chips will use AI to perform or accelerate machine learning tasks at the device level, creating more efficient robots and devices.
- Companies will increasingly deploy machine learning automation tools to accelerate data processing and insights, plus lower costs.
In Our Portfolio
AVG Portfolio Companies in the AI/Machine Learning Space
These are just a few of our portfolio companies advancing the tech of AI and machine learning:
- Dataminr: Leading real-time information discovery platform, leveraging AI/machine learning in detecting and disseminating critical events for businesses and governments
- Geopipe: Immersive 3D digital twins of the real world, built by machine learning, for autonomous vehicle and personnel training, gaming, and intelligence
- Pienso: Machine learning platform designed to facilitate, manipulate, and manage the interaction between algorithms and their data