Submitted by: Ankit Bhutani (Y9227094) Supervised by: Prof. Amitabha Mukerjee Prof. K S Venkatesh AUTO-ASSOCIATIVE NEURAL NETWORKS OUTPUT SIMILAR AS INPUT BOTTLENECK CONSTRAINT LINEAR ACTIVATION – PCA [Baldi et al., 1989] NON-LINEAR PCA [Kramer, 1991] – 5 layered network ALTERNATE SIGMOID AND LINEAR ACTIVATION EXTRACTS NON-LINEAR FACTORS ABILITY TO LEARN HIGHLY COMPLEX FUNCTIONS TACKLE THE NON-LINEAR STRUCTURE OF UNDERLYING DATA HEIRARCHICAL REPRESENTATION RESULTS FROM CIRCUIT THEORY – SINGLE LAYERED NETWORK WOULD NEED EXPONENTIALLY HIGH NUMBER OF HIDDEN UNITS DIFFICULTY IN TRAINING DEEP NETWORKS NON-CONVEX NATURE OF OPTIMIZATION GETS STUCK IN LOCAL MINIMA VANISHING OF GRADIENTS DURING BACKPROPAGATION SOLUTION -``INITIAL WEIGHTS MUST BE CLOSE TO A GOOD SOLUTION’’ – [Hinton et. al., 2006] GENERATIVE PRE-TRAINING FOLLOWED BY FINE-TUNING PRE-TRAINING INCREMENTAL LAYER-WISE TRAINING EACH LAYER ONLY TRIES TO REPRODUCE THE HIDDEN LAYER ACTIVATIONS OF PREVIOUS LAYER INITIALIZE THE AUTOENCODER WITH WEIGHTS LEARNT BY PRE-TRAINING PERFORM BACKPROPOAGATION AS USUAL STOCHASTIC – RESTRICTED BOLTZMANN MACHINES (RBMs) HIDDEN LAYER ACTIVATIONS (0-1) USED TO TAKE A PROBABILISTIC DECISION OF PUTTING 0 OR 1 MODEL LEARNS THE JOINT PROBABILITY OF 2 BINARY DISTRIBUTIONS - 1 IN INPUT AND THE OTHER IN HIDDEN LAYER EXACT METHODS – COMPUTATIONALLY INTRACTABLE NUMERICAL APPROXIMATION - CONTRASTIVE DIVERGENCE DETERMINISTIC – SHALLOW AUTOENCODERS HIDDEN LAYER ACTIVATIONS (0-1) ARE DIRECTLY USED FOR INPUT TO NEXT LAYER TRAINED BY BACKPROPAGATION DENOISING AUTOENCODERS CONTRACTIVE AUTOENCODERS SPARSE AUTOENCODERS TASK \ MODEL RBM SHALLOW AE CLASSIFIER [Hinton et al, 2006] and many others since then Investigated by [Bengio et al, 2007], [Ranzato et al, 2007], [Vincent et al, 2008], [Rifai et al, 2011] etc. DEEP AE [Hinton & Salakhutdinov, 2006] No significant results reported in literature - Gap MNIST Big and Small Digits Square & Room 2d Robot Arm 3d Robot Arm Libraries used Numpy, Scipy Theano – takes care of parallelization GPU Specifications Memory – 256 MB Frequency – 33 MHz Number of Cores – 240 Tesla C1060 REVERSE CROSS-ENTROPY X – Original input Z – Output Θ – Parameters – Weights and Biases RESULTS FROM PRELIMINARY EXPERIMENTS TIME TAKEN FOR TRAINING CONTRACTIVE AUTOENCODERS TAKE VERY LONG TO TRAIN EXPERIMENT USING SPARSE REPRESENTATIONS STRATEGY A – BOTTLENECK STRATEGY B – SPARSITY + BOTTLENECK STRATEGY C – NO CONSTRAINT + BOTTLENECK MOMENTUM INCORPORATING THE PREVIOUS UPDATE CANCELS OUT COMPONENTS IN OPPOSITE DIRECTIONS – PREVENTS OSCILLATION ADDS UP COMPONENTS IN SAME DIRECTION – SPEEDS UP TRAINING WEIGHT DECAY REGULARIZATION PREVENTS OVER-FITTING USING ALTERNATE LAYER SPARSITY WITH MOMENTUM & WEIGHT DECAY YIELDS BEST RESULTS MOTIVATION