1. Supernovas and Novel Insight: Where Machine Learning is Headed Next

    14:26

    from Andreessen Horowitz / Added

    As an astronomer at U.C. Berkeley Josh Bloom faces the problem every stargazer does, you can’t possibly scan the skies and catch everything. “There is so much data, that we just can’t have people looking at all of it,” Bloom says. Enter machine learning as an approximation of human intuition and cognition to aid in research and discovery -- not just in astronomy -- but also in physics, biology and other sciences. Bloom discusses how he is putting machine learning to work in his academic career, attacking as he puts it “a massive needle-in-the-haystack problem,” and how that experience has led to his own machine learning startup Wise.io.

    + More details
    • FLYR - FlyrLabs.com

      01:01

      from Alexander Mans / Added

      377 Plays / / 0 Comments

      FLYR was founded because erratic changes in airfares have long added an element of frustration that no traveler should ever have to deal with. Travel advice is everywhere, but only seems to speak in very general terms — there has never been a way to address each traveler's specific needs accurately, and on a large scale.

      + More details
      • Rithmio, The Gesture Recognition Platform

        02:19

        from Rithmio / Added

        76 Plays / / 0 Comments

        Rithmio’s gesture recognition platform integrates with motion-sensing devices at the sensor, OS or application-level to learn, track and analyze motion. Brands utilize Rithmio’s software to create accurate and personalized gesture-based products such as wearables, smartwatches or connected sporting equipment.

        + More details
        • Trivariate Function Graphic - Bike Ride

          00:17

          from Alexander Lorimer / Added

          5 Plays / / 0 Comments

          A machine learning algorithm has been used to develop an extremely complex trivariate mathematical function, modeled on 14 frames (spaced 1 second apart) taken from the video 'Cycling Safety Tips: Cyclists Visibility' by the Cycling Promotion Fund. The three input variables of the trivariate function are the x, y and t pixel co-ordinates of the 14 frames (with t representing time). The outputs of this function are the three RGB values of the respective pixel. Once the machine learning algorithm has created this continuous mathematical model of the sequence of images, it is possible to feed to trivariate function t co-ordinates that were not defined in the original sequence. The outputs of the function then act as interesting interpolations and extrapolations between and beyond the original frames. Here this technique has been used to increase the original 14 frames to 520 frames in total. More info here, http://www.aodlorimer.com

          + More details
          • Trivariate Function Graphic - Aurora

            00:20

            from Alexander Lorimer / Added

            5 Plays / / 0 Comments

            A machine learning algorithm has been used to develop an extremely complex trivariate mathematical function, modeled on 15 frames (spaced 1 second apart) taken from the video 'Aurora borealis timelapse - polar reflections autumn 2013' by the Antti Pietikäinen. The three input variables of the trivariate function are the x, y and t pixel co-ordinates of the 15 frames (with t representing time). The outputs of this function are the three RGB values of the respective pixel. Once the machine learning algorithm has created this continuous mathematical model of the sequence of images, it is possible to feed to trivariate function t co-ordinates that were not defined in the original sequence. The outputs of the function then act as interesting interpolations and extrapolations between and beyond the original frames. Here this technique has been used to increase the original 15 frames to 600 frames in total. More info here, http://www.aodlorimer.com

            + More details
            • Adgorithms - What We Do

              01:45

              from Albert Adgorithms / Added

              428 Plays / / 0 Comments

              Meet Albert, the first fully automated optimization manager. Albert manages digital campaigns by utilizing machine learning and algorithms to reach your desired KPI's.

              + More details
              • Deep Analytics

                02:53

                from OPT Data Science Lab / Added

                3 Plays / / 0 Comments

                + More details
                • Predicting Progression of Alzheimer's Disease with Machine Learning

                  04:00

                  from Josh Tycko / Added

                  2 Plays / / 0 Comments

                  Decision tree classifiers implemented to predict the progression of Alzheimer's Disease based on mixed data groups. Final Project for CIS 519: Machine Learning at the University of Pennsylvania. Music Credit: "Every Direction is North" El Ten Eleven

                  + More details
                  • CIS 519 Final Project

                    03:54

                    from Yongbo Qian / Added

                    4 Plays / / 0 Comments

                    Forecasting Demand of Bike Sharing Systems

                    + More details
                    • BroadMind Learns to Play Super Mario

                      07:46

                      from Dave Kotfis / Added

                      4 Plays / / 0 Comments

                      We have trained a neural network to learn to play Super Mario through practice and reinforcement by extending DeepMind's algorithm. http://deepmind.com http://www.cs.toronto.edu/~vmnih/docs/dqn.pdf What game should BroadMind tackle next?

                      + More details

                      What are Tags?

                      Tags

                      Tags are keywords that describe videos. For example, a video of your Hawaiian vacation might be tagged with "Hawaii," "beach," "surfing," and "sunburn."