Full search page →

Practice

Every interactive problem from the courses in one list: predict, run the lab, check yourself.

Study plans

Question

Showing 72 of 72

StatusQuestionChapterTypeLevel
Not startedImages as Arrays · Why does brightening make bright pixels dark?Add the same number to every pixel of a real photo and compare 8-bit wraparound with saturating and floating-point arithmetic.Images as ArraysImage labBeginner
Not startedML 01A · What does a 2 × 2 matrix do to a picture?Move the images of the two basis vectors and watch a real photograph follow. The determinant is the signed area scale, the singular values are the stretch along two perpendicular directions, and rank 1 flattens the picture onto a line.ML 01AImage labBeginner
Not startedIP 02 · What do you lose with fewer pixels, and what with fewer gray levels?Keep every s-th pixel and fewer gray levels of a real photo and of a synthetic zone plate, and see false patterns (aliasing) appear unless you filter first.Images as ArraysImage labBeginner
Not startedIP 04 · Does the sharpest low-pass filter give the best image?Filter a real photo in the frequency domain and compare an ideal, a Butterworth and a Gaussian low-pass at the same cutoff.IP 04Image labIntermediate
Not startedIP 05 · Why does undoing a blur amplify the noise?Blur a real photo with a Gaussian PSF, add a little noise, then divide by the blur in the frequency domain. See the plain inverse filter fail, and find the Wiener constant K that rescues it.IP 05Image labIntermediate
Not startedIP 06 · Why does brightening turn reds and oranges yellow, then white?Multiply a real color photo by a gain and compare three ways of handling the channels that overflow: clip each one, rescale the pixel, or tone-map it.IP 06Image labIntermediate
Not startedIP 08 · How many coefficients can you throw away?Transform a real photo with an 8 × 8 block DCT or a Haar wavelet, round every coefficient to a coarse step, and see where the lost information shows up.IP 07Image labIntermediate
Not startedIP 09 · Which shapes does a structuring element keep?Clean a thresholded photo of coins that carries specks, scratches and holes. Change the operation, the shape and the size of the structuring element, and see what each choice removes, fills or keeps.IP 09Image labIntermediate
Not startedIP 10 · Where does one global threshold fail?Dim one side of a page and watch Otsu’s single threshold lose the text, while a threshold that follows the local light keeps it.IP 10Image labAdvanced
Not startedIP 11 · Does the snake find the edge, or get stuck?Run a real snake, the Kass–Witkin–Terzopoulos contour of IP 11, on a photograph of coins: start a circle around one coin and see how far from its rim the circle may begin before the snake settles in the wrong place.IP 11Image labAdvanced
Not startedIP 12 · Probe a pixel: flat, edge or corner?Slide a small window over a real photograph and watch the window’s gradient vectors, its structure-tensor ellipse and the Harris response decide whether the spot is flat, an edge or a corner.IP 12Image labAdvanced
Not startedCount the next wordBuild an n-gram distribution from a tiny editable corpus. Inspect counts, sentence boundaries and smoothing.NLP 01Interactive plotBeginner
Not startedBuild subwords one merge at a timeTrain byte-pair encoding on a tiny weighted vocabulary and apply the learned merge order to a new word.NLP 01Interactive plotBeginner
Not startedDoes “everyone read a book” mean the same book?Change a small reading relation and evaluate two quantifier orders exactly.NLP 02Interactive plotBeginner
Not startedDoes padding change a mean-pooled sentence?Pad a three-word sentence with zero vectors and compare a correct masked mean with a mean that counts the padding in its denominator.NLP 03Interactive plotBeginner
Not startedLocal tag choices versus a coherent sequenceCombine emission scores, transition scores and BIO span constraints to choose a complete tag path.NLP 06Interactive plotIntermediate
Not startedWhat may a masked predictor see?Change a right-context clue, hide it causally, then attach different output heads to the same small contextual vector.NLP 07Interactive plotIntermediate
Not startedTemperature, top-k and top-p change the distributionFollow fixed logits through temperature scaling, top-k truncation, nucleus truncation and renormalization.NLP 08Interactive plotIntermediate
Not startedWhy does this document rank first?Inspect BM25 term saturation and document-length normalization on a fully visible local corpus.NLP 09Interactive plotAdvanced
Not startedPR 04 · Does changing a feature’s unit change its neighbours?Run leave-one-out k-nearest-neighbour classification on the real iris measurements, then change the unit of one feature: no flower moves, yet the neighbours and the accuracy do.IP 13Image labAdvanced
Not startedWhen should a classifier change its mind?Move a decision boundary by changing class frequency and the cost of a false alarm.PR 01Interactive plotBeginner
Not startedPR 02 · How wide should a density kernel be?Change the bandwidth and inspect individual kernels, the estimated density and leave-one-out prediction.PR 02Interactive plotIntermediate
Not startedPR 03 · Complexity versus shrinkageFit polynomial basis functions with ridge regression and separate training error from recovery of the underlying function.PR 03Interactive plotIntermediate
Not startedPR 05 · Does maximum variance preserve the classes?Rotate a one-dimensional projection and compare reconstruction error with class separation.PR 05Interactive plotIntermediate
Not startedPR 06 · Match shape despite timing changesWarp a signal in time and inspect a dynamic time warping cost matrix and its optimal alignment path.PR 06Interactive plotAdvanced
Not startedPR 07 · Watch soft assignments move the meansStep through Gaussian-mixture EM and observe responsibilities, likelihood and the symmetric-initialization trap.PR 07Interactive plotAdvanced
Not startedPR 08 · Which direction of KL are you minimizing?Move and widen a Gaussian approximation to a two-mode target; compare penalties for missing mass and filling the valley.PR 08Interactive plotAdvanced
Not startedPR 09 · Does averaging many unstable trees beat one tree?Grow real decision trees on 300 noisy two-moons points, bootstrap them, let them vote, and watch the test error fall and then flatten as trees are added.PR 09Image labAdvanced
Not startedPR 10 · Does confidence survive a shifted population?Set a rejection threshold, shift both class distributions and compare accepted coverage with selective error.PR 10Interactive plotAdvanced
Not startedTwo hidden units can bend a decision regionCompare a fixed two-unit network with and without nonlinear activations on the XOR corners.NLP 03Interactive plotBeginner
Not startedSame predictions, different probability qualityApply one positive temperature to fixed validation logits and compare accuracy, log loss and squared probability error.NLP 03Interactive plotBeginner
Not startedAlignment, hard negatives and collapsed featuresMove paired and unpaired unit vectors, then inspect the normalized matching loss used by contrastive learning.NLP 04Interactive plotBeginner
Not startedSelective state updates: remember the useful inputsCompare a fixed recurrent decay with input-dependent write decisions while distractors pass through a sequence.NLP 05Interactive plotIntermediate
Not startedAttention: which tokens are visible?Compare bidirectional and causal attention using the same query, keys and values.NLP 07Interactive plotIntermediate
Not startedFactor a tiny image into conditional predictionsEnumerate all sixteen binary 2×2 images under a first-order pixel model, then change the generation order.NLP 09Interactive plotAdvanced
Not startedStretch space, change density: the Jacobian mattersPush a Gaussian through an invertible nonlinear map and preserve probability mass with its derivative.ML 01BInteractive plotBeginner
Not startedWhy a smaller finite-difference step can be worseCheck the derivative of exp(x), separating truncation error from floating-point cancellation.ML 01CInteractive plotBeginner
Not startedA correct velocity field still needs a numerical solverTransport particles through a known velocity field and compare integration error at equal step counts.ML 01CInteractive plotBeginner
Not startedML 02 · Can a tiny network learn a curved boundary?Train a network with 2 inputs, m tanh hidden units and 1 sigmoid output on the two moons and the two circles, one gradient-descent step at a time. Watch the decision map and the training and test loss: too few hidden units, too small a learning rate and too large a learning rate each fail in their own way.ML 02Image labIntermediate
Not startedA VAE trades reconstruction against prior agreementMove and widen a Gaussian posterior, then compute expected decoder loss and the exact KL penalty.ML 06AInteractive plotIntermediate
Not startedML 08 · What does a network see when it predicts noise, the image or a velocity?Mix the astronaut photograph with fixed noise, look at the noise, clean-image and velocity targets side by side, then give the network a small error and see which conversion back to the clean image survives.ML 08Image labAdvanced
Not startedDLCV 01 · How much of the image can one output unit see?Pick a stack of convolution layers and a spot in a real photograph: the exact box of input pixels that one output unit depends on is drawn on the image, with its size, jump and the share of the image it covers.ML 03Image labIntermediate
Not startedDLCV 02 · A decision threshold changes which mistakes you makeMove the decision threshold on fixed classifier scores. The confusion matrix changes, and the operating point slides along the ROC and precision-recall curves built from the same scores.DLCV 02Interactive plotAdvanced
Not startedBoxes, overlap and suppressionWhen do overlapping detections represent one object? Inspect how greedy suppression makes that decision.DLCV 03Interactive plotAdvanced
Not startedDLCV 03B · Is a small yaw error harmless?Turn a predicted box a few degrees away from the ground truth in the bird’s-eye view and watch the exact rotated IoU fall while the centre distance stays at zero.DLCV 03BImage labAdvanced
Not startedDLCV 04 · Does high pixel accuracy mean a good segmentation?Simulate a segmentation network that predicts at a coarse output stride and compare pixel accuracy, mean IoU and the IoU of the smallest class.DLCV 04Image labAdvanced
Not startedDLCV 05 · Is the argmax of a heatmap the keypoint?Render a Gaussian heatmap at output stride 1, 2, 4 or 8 for 500 random sub-pixel keypoints and compare three ways to read the position back: the brightest cell, a quarter-cell offset and a Taylor step.DLCV 05Image labAdvanced
Not startedDLCV 06 · How does CTC write a double letter?Edit the blank probability in the frames between two equal letters of a toy recognizer output, and watch greedy best-path decoding and prefix beam search read it.DLCV 06Image labAdvanced
Not startedDLCV 07 · Why do identities swap when objects meet?Track three coloured blocks through 30 synthetic frames with a Kalman filter, a cost matrix and the Hungarian algorithm, and watch two identities swap when only box overlap is used.DLCV 07Image labAdvanced
Not startedDLCV 04 · A sharp guess can be less faithfulA blurry observation fits two sharp clean signals. The blurry average has the lowest expected squared error; a sharp guess looks better but is usually farther from the truth.DLCV 09Interactive plotAdvanced
Not startedDLCV 09C · Gain cannot create missing photonsSimulate a dim photo with photon and read noise, then compare more light, more digital gain and a denoiser.DLCV 09CImage labAdvanced
Not startedDLCV 09C · How big must a 3D LUT be?Bake one smooth color enhancement into an M × M × M table and apply it with a nearest-node lookup or with trilinear interpolation. Compare both with the exact transform.DLCV 09CImage labAdvanced
Not startedDLCV 10 · Does averaging two photos fuse them?Two shots of one scene are each sharp in a different region. Average them, or pick every pixel from the shot with more local detail, and score both against the all-in-focus truth.DLCV 10Image labAdvanced
Not startedDLCV 11A · Is a bigger matching window always better?Run block matching on a synthetic stereo pair with known disparity, and see small windows turn noisy while large windows fatten the foreground.DLCV 11AImage labAdvanced
Not startedDLCV 07 · Small dynamics errors compound in rolloutsCompare one-step predictions, open-loop rollout and periodic observation refresh under the same actions.DLCV 13Interactive plotAdvanced
Not startedDLCV 14 · Why can an invisible change flip a prediction?Train a linear classifier on noisy stripe images of 64, 256 or 1,024 pixels, then change every pixel by the same few levels, once along the loss gradient (FGSM) and once with random signs.DLCV 14Image labAdvanced
Not startedDoes the whole segment stay feasible?Place two feasible points on a circle, slide along the segment between them, and compare a disk with an annulus.OPT 01Interactive plotBeginner
Not startedCompare a function with its chordMeasure the Jensen gap while moving between two inputs. A negative gap supplies a concrete counterexample.OPT 02Interactive plotIntermediate
Not startedWhy clipping a solution can be wrongCompare coordinate clipping with the exact solution of a coupled quadratic under nonnegative constraints.OPT 03Interactive plotIntermediate
Not startedBuild a primal–dual certificateChange a candidate point and multiplier. A small gap is meaningful only with feasibility.OPT 04Interactive plotAdvanced
Not startedOne outlier, three meanings of best fitMove one observation and compare squared, absolute and Huber losses for fitting a constant.OPT 05Interactive plotAdvanced
Not startedFollow a gradient down a narrow valleySee why a step size that works on a round bowl can diverge on a narrow one.OPT 06Interactive plotAdvanced
Not startedFollow the central path to the optimal vertexRaise the barrier weight t on a two-variable linear program and watch the central point, its dual bound and the gap m/t.OPT 07Interactive plotAdvanced
Not startedWhy a proximal step creates exact zerosCompare soft thresholding with ordinary shrinkage and inspect its nonsmooth objective.OPT 08Interactive plotAdvanced
Not startedSeparate entropy from model mismatchChange the true probability and the model probability; measure the extra coding cost caused by a wrong model.ML 01BInteractive plotBeginner
Not startedHow much information survives noise?Separate source uncertainty, transmitted information and channel capacity.IT 01Interactive plotBeginner
Not startedA likely block is not a typical setCompare the probability mass of a typical set with its fraction of all possible binary blocks.IT 02Interactive plotIntermediate
Not startedGive common symbols shorter codesBuild a binary prefix code and compare its average length with source entropy.IT 03Interactive plotIntermediate
Not startedAllocate power where it carries more informationCompare equal allocation with water-filling across two parallel Gaussian channels.IT 05Interactive plotAdvanced
Not startedHow much rate buys a given error?Inspect the Bernoulli rate–distortion curve and an attaining joint distribution under Hamming error.IT 06Interactive plotAdvanced
Not startedTrade false alarms for missed alternativesSlide a decision threshold across two exact binomial distributions and compute both error probabilities.IT 07Interactive plotAdvanced
Not startedTwo useless clues can be useful togetherCompute information about a noisy XOR label from each independent bit and from their pair.IT 08Interactive plotAdvanced