Mila > Industry > AI Activation Program
INDUSTRY

AI Activation Program

Helping small and medium-sized businesses begin their AI adoption journey.

Helping SMEs kick-start the adoption of advanced machine learning

SME managers often encounter barriers to entry that delay the benefits promised by artificial intelligence (AI) for their organizations. There is often a significant gap between the SME's starting point in AI and the advance use of these technologies.

Our AI Activation program equips SMEs with the necessary tools to bridge this gap, whether they have teams of experienced data scientists or rely on limited resources.

Contact us
PLAY VIDEO

Our Services

Through advisory engagements lasting a few months, we guide organizations towards the advance use of AI to improve their productivity, generating positive impacts on society.

Our team can offer general or customized demonstrations of ways to integrate AI into your operations:


Anomaly Detection

Detect production anomalies using computer vision

Biomedecine

Increase the efficiency of biomedical experimental protocol steps through various AI techniques

Support for Education

Support the education of young students through automatic text classification using natural language processing (NLP) techniques or recommendation systems

Climate Change

Identify events that may contribute to climate change using computer vision with satellite images

Recyclable Materials

Improve sorting of recyclable materials with computer vision

Agriculture

Precision agriculture based on computer vision and robotics

Energy Efficiency

Increase energy efficiency using various AI applications and techniques

Other

Many other business sectors can benefit from AI. Contact us to learn more.

Recent Projects

Alloprof

Alloprof: Helping elementary and high school students with a tool that offers instant explanations to their school-related questions

Alloprof: Helping elementary and high school students with a tool that offers instant explanations to their school-related questions

Alloprof, a Quebec nonprofit organization offering professional and stimulating services free of charge, helps students in Quebec excel by transforming school challenges into successes. With Alloprof's Help Zone, students can ask a question about a school subject, obtain an explanation from another user or an Alloprof moderator, and be given suggestions for educational concept sheets to read.To prevent situations where students get stuck on a question and give up on their study sessions, it is vital that they be helped quickly so that they can be encouraged to persevere. Thanks to Artificial Intelligence (AI), a student seeking help in the Help Zone receives assistance adapted to their needs, instantly.In the summer of 2022, Alloprof and Mila leveraged modern Deep Learning (DL) techniques to improve the Help Zone. The new version of the tool has significantly improved the relevance of the content offered and the speed at which students are able to access and consult the concept sheets containing the explanations they need.The collaborative work spanned four months. It began with discussions aimed at understanding the problem and identifying the available data for training the DL models. Meticulous work then made it possible to experiment with varied approaches, measure and compare the results, and draw a conclusion with which the team could define the model capable of meeting Alloprof’s needs and quality standards. In the last stage of the project, Mila provided the organization with a number of knowledge transfer documents, the trained model, and the entire code associated with the project. Now, in the space of half a second, the model can suggest relevant content to a student asking a question by offering up to three concept sheets correlated to that student’s level of education. In over 50% of cases, at least one of these sheets is recommended in the explanation provided by a moderator later on. More broadly, this AI application will provide better support for Quebec’s students in their study sessions and ultimately in their success in school.

Download the scientific report (in French)
Dubois Agrinovation

Dubois Agrinovation : improving employee experience and efficiency in the agri-food sector

Dubois Agrinovation : improving employee experience and efficiency in the agri-food sector

Dubois Agrinovation is an innovative supplier of agricultural equipment, providing solutions for customized irrigation, plasticulture and protective equipment. The organization offers a vast catalog of irrigation system parts, which are listed on the company’s website. In the summer of 2022, Dubois Agrinovation called on Mila’s AI Activation team with the aim of rethinking their approach to a very repetitive and tedious task for their employees: Identifying replacement parts that match their customers’ needs based on photos of the item to be replaced.

The Mila team conducted a demonstration to illustrate the potential for optimizing the business of selling parts for farm irrigation systems.

The computer vision-based system developed by Mila using deep learning techniques was used to demonstrate the potential of AI to Dubois Agrinovation employees. It can drastically speed up the process of identifying the spare parts needed by the company’s customers: This task, which otherwise takes up to 30 minutes, can now be achieved in a few seconds. To accomplish this, the system only needs a photo of the customer’s part and a photo of the same part from the catalog to train the model. This technology has many other benefits apart from the improved efficiency. The company’s irrigation system experts can now focus on more interesting tasks that require them to use their critical thinking skills. Once rolled out, the new process also allows the company’s customers, or its less experienced employees, to perform this task independently.

The collaborative project took place over four months during which Mila and Dubois Agrinovation identified the problem and the relevant data for training different models to help classify images of items and parts of equipment in the catalog. In its current state, the selected model, once trained on Dubois Agrinovation’s data, allows a short list of parts to be generated from the catalog that are similar to the image of the customer’s used part. Based on the data collected, it was found that the required part was listed in 80% of the cases. All the work carried out was shared with the company and this AI solution will allow it to improve its productivity in the management of parts sales and ultimately to improve its customer experience.

Meet the Team

Applied Machine Learning Scientist
Principal Advisor, Catalytic Cluster
Applied Machine Learning Scientist
Applied Machine Learning Scientist
Principal Advisor, Catalytic Cluster
Applied Machine Learning Scientist
Applied Machine Learning Scientist
Principal Advisor, AI Adoption
Principal Advisor, AI Adoption
Principal Advisor, AI Adoption
Applied Machine Learning Scientist
Applied Machine Learning Scientist
Join us

Contact our AI experts

Contact us