Machine learning company Maluuba, with headquarters in Waterloo, Ontario and a research office in Montreal, has applied an algorithm to the text of J.K. Rowling’s bestselling novel Harry Potter and the Philosopher’s Stone, along with several hundred other children’s stories, to read text in such a way that it can then answer questions afterward.
Maluuba has also just announced the opening of an R&D lab in Montreal, staffed by Yoshua Bengio of the Université de Montréal’s Montreal Institute for Learning Algorithms (MILA) in partnership with reinforcement learning expert Richard Sutton from the Alberta Innovates Centre for Machine Learning, to make advances in the fields of Natural Language Understanding (NLU) and artificial intelligence (AI).
Taking a deep learning approach, Maluuba trained its algorithm to approach the Harry Potter text from several levels of textual abstraction, word, sentence, paragraph, etc.
And while a certain contingent of tech utopians may very well look at Maluuba’s case study as the smoking gun they need for shutting down Humanities departments in universities everywhere, the company itself makes clear that using an algorithm to comprehend literature is a stepping stone to more practical uses.
“For a computer to understand humans speaking in natural language and respond appropriately, it needs to capture and represent a large amount of knowledge that is not just words, but also common sense and context about the topic being discussed by the human,” said Maluuba cofounder & CEO Sam Pasupalak. “Maluuba is working with leading experts and the world’s premiere academic center for deep learning to design systems that can represent knowledge and answer questions in natural language. The potential applications of this research are staggering.”
Maluuba’s approach improved by 8% on the previous existing word-matching technique applied to the MCTest of 72%, reaching 80% accuracy, and improved by 3% the previous benchmark for deep learning techniques, the DSTC2, reaching 83%, up from 80%.
Maluuba scientific advisor Bengio calls that “a big jump”, but also cautions that machines still have a long way to go before putting human consciousness out of a job.
“While we’re closer to the goal of getting machines to exercise reasoning and understand conversational language, we still have a long way to go,” said Bengio. “Maluuba has made great strides with its contributions to the field of machine learning and NLP. Their long-term vision and focus on truly perfecting the methodology is refreshing and makes me confident that we’ll see exciting tech and research advancements from them this year.”
Founded in 2011 by a group of University of Waterloo graduates to create “Siri for Android”, this past January, Maluuba closed a $9 million Series A round of funding, from Montreal-based Emerillon Capital, San Francisco-based Nautilus Ventures, and undisclosed strategic investors.
This article is brought to you by VersaPay (TSXV:VPY) . VersaPay ARC allows you to manage AR with no paper, no hassle, and no trips to the bank. This not only allows you to eliminate costs and get paid faster, but improves your customers’ experience doing business with you. VersaPay ARC is the simple and secure way to manage your entire AR process online. Our cloud-based solution allows you to collaborate with your customers through your own web portal. Click here to learn more.
The company had previously raised a $2 million seed round from Samsung Ventures in 2012.
Maluuba today enables interactive natural language and conversational dialogue experiences in over 50 million smart devices globally including IoT, mobile phones and smart TVs for several industry OEMs including LG, supporting more than 10 languages.
The company is also working with Qualcomm to integrate the voice command system in Maserati cars with Qualcomm’s connected speaker systems.
J.K. Rowling would probably not be much impressed to hear that a company has created a piece of software that might potentially be touted as something that can “read Harry Potter so that you don’t have to.” But she’s made her money. It won’t likely keep her awake at night.
What might bother her, though, is that while a piece of software may very well be able to respond to the experience of reading in a way that could save a human being the headache of parsing a technical manual for a newly purchased deep fryer or home theatre system, that same piece of software will probably never seriously communicate anything about the essentially embodied experience of reading literature.
Reading In Search of Lost Time by Marcel Proust is not a waste of anyone’s time, having at least as much to do with experiencing the duration of inhabiting the text than can be summed up by a bullet point summary.
That said, at this point, I would probably pay a machine to tell me whether or not there’s any point to carving out a year of my life to read Karl Ove Knausgård (“Scandinavian man has children, is unhappy, gets drunk.” Thanks, computer!)