Deep Neural Networks (DNN) are computationally intensive to train. It consists of a large number of multidimensional dot products between many weights and input vectors. However, there can be significant similarities among input vectors. If one input vector is similar to another, its computations with the weights are similar to those of the other and, therefore, can be skipped by reusing the already-computed results. We propose a novel scheme, called MERCURY, to exploit input similarity during DNN training in a hardware accelerator. MERCURY uses Random Projection with Quantization (RPQ) to convert an input vector to a bit sequence, called Signature. A cache (MCACHE) stores signatures of recent input vectors along with the computed results. If the Signature of a new input vector matches that of an already existing vector in the MCACHE, the two vectors are found to have similarities. Therefore, the already-computed result is reused for the new vector. To the best of our knowledge, MERCURY is the first work that exploits input similarity using RPQ for accelerating DNN training in hardware. The paper presents a detailed design, workflow, and implementation of the MERCURY. Our experimental evaluation with twelve different deep learning models shows that MERCURY saves a significant number of computations and speeds up the model training by an average of 1.97× with an accuracy similar to the baseline system.

A handful of research has demonstrated the potential of mouse tracking for emotion assessment. Yet, theoretical and empirical bases for this approach remain opaque. If emotion influences motor control (e.g., controlling the movement of the computer mouse), how does that happen? What experimental situations constrain or promote the purported connection? To address these questions, we examined how prior emotional experience (viewing emotional photos) would influence participants' motor activity, as measured by the movement of the computer cursor. Results from two experiments indicate that emotional experience impacts both temporal (peak velocity) and spatial characteristics (deviation of the trajectory) of the cursor motion. But there are clear gender differences; for male participants, emotion influenced temporal features (peak velocity) but this impact was absent in female participants. It is suggested that emotion intervenes motor vigor and decision-making processes differently in male and female participants. Theoretical rationales and design factors that are essential for trajectory-based emotion assessment are also discussed.