Over the last few weeks we’ve been working on applying Deep Learning algorithms for a new VentureRadar feature we’re adding in the coming weeks. This piqued my interest in finding out more about the startups leading the way in developing and applying Deep Learning, so I decided to pick out the eighteen highest ranked companies in this emerging field from the VentureRadar database, and take a closer look at them.
You can also search VentureRadar for “Deep Learning” to find out about more companies in this exciting area.
You can find out more about each company in the profiles below.
Enlitic uses deep learning and image analysis to help doctors make diagnoses and spot abnormalities in medical images. For example, Enlitic can analyze medical images such as X-rays, MRIs, or CT scans for trends in the data or anomalies in individual images.
Synapsify builds applications that semantically read and learn from written content similar to humans, for accelerated discovery, insight and recommendations. The company’s vision is to allow anyone to apply and benefit from machine intelligence without the need for technical expertise or resources, resulting in true actionable insight and discovery.
Affectiva’s technology employs sophisticated computer vision algorithms to capture and identify emotion reactions to visual stimulus. Affdex, Affectiva’s flagship product, is easy to use, anywhere–only a webcam is required and there is no software to install. And Affdex is unobtrusive–there are no messy wires or electrodes.
Ripjar helps businesses exploit big data. The Ripjar application suite allows organisations to bring together targeted external data (social media, news feeds, blogs and internet pages) and internal information to expose unique insights allowing them to make timely business critical decisions. The Ripjar product set allows to monitor, analyse and explore blended datasets in a coherent manner.
Deepomatic is building a smart button to link any desirable product in media images to the same or similar product available on e-commerce sites. Publishers get their images scanned by algorithms that detect and identify inspiring products (e.g. fashion products). By combining the automatic understanding of product attributes and the comparison of patterns and colors, Deepomatic links those images to e-commerce products which are the same or very similar. Deepomatic shares the new stream of revenues with the media owners.
Indico develops machine learning models that automate understanding, with the goal of making every programmer a 10x data scientist. Each of the company’s models performs a unique task. For example, its “Text Tags” model determines the topic of paragraph, and the “Facial Localization” model finds all the faces in an image. Users can combine models to answer complex questions about their data.
Clarifai’s first image recognition systems held the top 5 spots for classifying objects in images in the ImageNet 2013 competition. Since then Clarifai’s deep learning systems have improved orders of magnitude in speed, vocabulary size, memory footprint and have expanded beyond images to extract knowledge from all forms of data. The hub of Clarifai’s technology is a high performance deep learning API on which a new generation of intelligent applications is being built. It enables Clarifai to combat everyday problems with high tech solutions by providing powerful machine learning systems to everyone in new and innovative ways.
Atomwise uses Deep Learning Neural Networks to discover new medicines. The company says it achieves the world’s best results for new drug hit discovery, binding affinity prediction, and toxicity detection. Atomwise predicts drug candidates for pharmaceutical companies, startups, and research institutions.
Descartes Labs is teaching computers how to see the world and how it changes over time based on Deep Learning and advanced remote sensing algorithms. Their first application is to use massive amounts of satellite imagery, across both visible and non-visible spectrums, to gain a better understanding of global crop production.
MetaMind wants to make Deep Learning accessible to everyone; the company is building an AI platform for natural language processing, image understanding, and knowledge base analytics. The company offers products for medical imaging, food recognition, and custom solutions.
Quantified Skin is an artificial-intelligence-based platform that learns, adapts and recommends activities to improve a user’s health. Quantified Skin aims at reducing chronic diseases due to obesity, which is America’s greatest epidemic. Preliminary results have shown behavior modification in users towards healthier lifestyles while using Quantified Skin’s platform.
Deep Genomics has world-leading expertise in machine learning, genome biology and precision medicine. The company is inventing a new generation of computational technologies that can tell what will happen within a cell when DNA is altered by genetic variation, whether natural or therapeutic.
HyperVerge uses deep learning algorithms to process consumer images and videos on the cloud. Proprietary and patented image technology modules developed by HyperVerge for processing of images include: face detection; face recognition; scene recognition; bad photo detection; duplicate photo detection; photo clustering; photo album summarization; face enhancement; photo enhanacement.
Idibon’s cloud-based natural language processing services enable organizations to organize and structure written language to answer critical business questions and automate their processes. Idibon has an accurate and adaptable systems and support NER/text extraction, sentiment analysis, text/content categorization, language detection/identification.
Tractable / Computer Vision
Tractable is developing proprietary machine learning algorithms, with a focus on deep learning for computer vision. The company’s focus is getting unlabelled data and supervised learning to work together. Application areas include insurance claims, industrial inspection; remote monitoring, among others.
Trustingsocial is inventing consumer credit rating for emerging markets by applying Big Data and Deep Learning technologies to social, mobile and web data. Its scoring algorithm learns from vast social datasets to predict short and long-term income and creditworthiness. It complements the FICO score, which is based on a customerís credit history. By leveraging social network data, the scoring system is applicable to billions users worldwide.
Trak.io is a cloud-based platform for Software-As-A-Service (SaaS) companies to track their customer data. Combining an efficient System Of Record process with deep learning & pattern recognition, the company’s fully managed platform builds predictive models of customer behaviour. Trak.io takes in all of a clients customer data, such as feature usage, payments, support tickets and email history, and then automatically segments users based on that data.
Skymind is a business intelligence and enterprise software firm that analyzes media, image and sound to locate and quantify patterns that impact businesses. The company is comprised of a team of deep-learning specialists and semi-sentient robots, and the creators of the world’s first open-source, distributed, commercial-grade deep-learning framework: Deeplearning4j.org.