backdoor attack
interpretability
Backdoor Attack is a training-time attack that implants hidden triggers causing targeted model misbehavior - It preserves normal accuracy while enabling attacker-controlled…
29,085 technical terms and definitions
interpretability
Backdoor Attack is a training-time attack that implants hidden triggers causing targeted model misbehavior - It preserves normal accuracy while enabling attacker-controlled…
Backdoor Attack Detection Introduction & Motivation Backdoor Attack Detection addresses a central problem in robustness of advrobust models under adversarial conditions: how…
Backdoor Attack Detection (Extended Study) Introduction & Motivation Backdoor Attack Detection (Extended Study) addresses a central problem in robustness of advrobust models under…
Backdoor Attack Detection (Extended Study) Introduction & Motivation Backdoor Attack Detection (Extended Study) addresses a central problem in robustness of advrobust models under…
ai safety
Backdoor Attacks are a class of adversarial attacks where an attacker embeds a hidden trigger pattern in the model during training — the model behaves normally on clean inputs but…
backdoor detection, detection defense, ai cifar, backdoor detection & defense
Backdoor Detection & Defense Introduction & Motivation Backdoor attacks: inject malicious behavior via data poisoning. Stealthy attacks on training. Applications: security, threat…
backdoor insertion detection rate prediction, backdoor insertion detection, insertion detection rate, detection rate prediction, backdoor insertion, insertion detection
Backdoor Insertion Detection Rate Prediction with Machine Learning Introduction & Motivation Backdoor Insertion Detection Rate Prediction with Machine Learning addresses a central…
backend chip process flow, back end of line, BEOL, dual damascene, copper interconnect, chip wiring stack, how chips are wired
The backend of a chip is everything built after the transistors already work — ten to fifteen levels of copper wiring, stacked one on top of the other, whose entire job is…
copper interconnect damascene, low k dielectric, via contact metal, multilayer wiring
Copper dual damascene interconnect architectures, electrochemical superfilling, and barrier-seed metallization constitute the back-end-of-line (BEOL) wiring systems that route…
infrastructure
Backfill scheduling is the opportunistic scheduler strategy that runs smaller jobs in temporary gaps without delaying higher-priority reservations - it increases cluster…
predicting backgate coupling capacitance in multi-gate devices, predicting backgate coupling, backgate coupling capacitance, coupling capacitance multi-gate, predicting backgate
Predicting backgate coupling capacitance in multi-gate devices Introduction & Motivation Predicting backgate coupling capacitance in multi-gate devices addresses a central problem…
backgrind subsurface damage depth prediction, backgrind subsurface damage, subsurface damage depth, damage depth prediction, backgrind subsurface, subsurface damage
Backgrind Subsurface Damage Depth Prediction with Machine Learning Introduction & Motivation Backgrind Subsurface Damage Depth Prediction with Machine Learning addresses a central…
backgrind tape delamination prediction, backgrind tape delamination, tape delamination prediction, backgrind tape, tape delamination, delamination prediction
Backgrind Tape Delamination Prediction with Machine Learning Introduction & Motivation Backgrind Tape Delamination Prediction with Machine Learning addresses a central problem…
backgrind tape residue prediction machine learning, backgrind tape, residue prediction machine, backgrind tape residue prediction, backgrind tape residue, tape residue prediction
Backgrind Tape Residue Prediction with Machine Learning Introduction & Motivation Backgrind Tape Residue Prediction with Machine Learning addresses a central problem in wafer…
backgrind wheel dressing wear prediction machine learning, backgrind wheel, dressing wear prediction, backgrind wheel dressing wear prediction, backgrind wheel dressing
Backgrind Wheel Dressing Wear Prediction with Machine Learning Introduction & Motivation Backgrind Wheel Dressing Wear Prediction with Machine Learning addresses a central problem…
computer vision
Background Bias is the tendency of image classifiers to rely on background context for classification instead of the actual object — the model learns to associate specific…
video understanding
Background modeling is the process of statistically representing per-pixel scene appearance over time so moving foreground can be separated from repetitive or changing background…
metrology
Background Signal is the baseline signal detected by an instrument in the absence of the target analyte — arising from detector noise, stray light, contamination, matrix emission…
video understanding
Background subtraction is the classical motion detection technique that models static scene appearance and flags pixels that deviate from that model as foreground activity - it is…
supply chain & logistics
Backorder is an unfulfilled order quantity recorded for later shipment when inventory becomes available - It provides continuity of demand capture but signals supply imbalance.…
backprop, chain rule, gradient computation
Backpropagation is the algorithm that lets a neural network learn. After the network makes a prediction and we measure how wrong it was, backpropagation efficiently computes how…
automatic differentiation computation graph, gradient checkpointing memory tradeoffs, vanishing exploding gradient mitigation, optimizer gradient flow diagnostics
Backpropagation Gradient Chain Rule is the optimization backbone of modern deep learning, enabling efficient parameter updates by propagating loss sensitivity from outputs to all…
backpropagation gradient flow vanishing exploding gradients, backpropagation gradient, flow vanishing exploding, mnist passed, backpropagation gradient flow, gradient flow
Backpropagation & Gradient Flow: Vanishing and Exploding Gradients Introduction & Motivation Backpropagation: compute gradients via chain rule; foundation of deep learning.…
optimization
Backpropagation Through Time (BPTT) is the standard algorithm for computing gradients in recurrent neural networks — unrolling the recurrent computation through time steps…
backscatter electron detector calibration prediction machine learning, backscatter electron, detector calibration prediction, sem, backscatter electron detector
Backscatter Electron Detector Calibration Prediction with Machine Learning Introduction & Motivation Backscatter Electron Detector Calibration Prediction with Machine Learning…
Power Delivery Network, BSPDN, interconnect, buried power rail, hbm4 power delivery network, hbm power tsv, signal tsv vs power tsv
Backside power delivery networks (BSPDN) move the power distribution grid from the front side of the wafer, where it competes with signal routing, to the back side, where it…
process
Backside alignment is the lithography alignment method that registers backside process patterns to frontside device features through wafer-thickness references - it enables…
process
Backside contact formation is the process of landing an electrical contact on the underside of a transistor after the wafer has been flipped and thinned. It is the step that makes…
process
Backside Damage Gettering is a simple extrinsic gettering technique that introduces mechanical damage (scratches, abrasion, microcracks) on the non-active backside of the wafer…
process
Backside damage removal is the post-grinding process that eliminates stressed or cracked silicon layers from the wafer rear surface - it restores surface integrity before…
backside deposition removal drift prediction, backside deposition removal, deposition removal drift, removal drift prediction, backside deposition, deposition removal
Backside Deposition Removal Drift Prediction with Machine Learning Introduction & Motivation Backside Deposition Removal Drift Prediction with Machine Learning addresses a central…
cvd backside gas, wafer backside gas, helium backside gas, helium backside cooling, wafer backside cooling, esc backside gas, backside gas pressure, backside helium leak, backside gas thermal contact, multi zone backside gas, heat transfer gas wafer
Backside gas is a controlled low-pressure heat-transfer medium introduced into the microscopic gap between a process wafer and its chuck or pedestal. Helium is common because it…
production
Backside grinding (wafer thinning) reduces wafer thickness from the standard 775μm (300mm wafer) to 50-200μm by mechanically grinding the wafer backside after front-side device…
process
Backside grinding is the mechanical thinning process that removes silicon from the wafer rear surface to reach target thickness for packaging - it is the primary material-removal…
backside helium leak rate prediction machine learning, backside helium, leak rate prediction, backside helium leak rate prediction, backside helium leak, helium leak rate
Backside Helium Leak Rate Prediction with Machine Learning Introduction & Motivation Backside Helium Leak Rate Prediction with Machine Learning addresses a central problem…
backside helium thermal contact drift prediction, backside helium thermal, helium thermal contact, thermal contact drift, contact drift prediction, backside helium
Backside Helium Thermal Contact Drift Prediction with Machine Learning Introduction & Motivation Backside Helium Thermal Contact Drift Prediction with Machine Learning addresses…
backside illuminated sensor wafer thinning uniformity prediction with machine learning, backside-illuminated sensor wafer thinning uniformity prediction
Backside-Illuminated Sensor Wafer Thinning Uniformity Prediction with Machine Learning Introduction & Motivation Backside-Illuminated Sensor Wafer Thinning Uniformity Prediction…
bsi sensor, bsi cmos image sensor, backside illuminated, bsi technology
Backside Illumination (BSI) is the CMOS image sensor architecture where light enters from the back of the silicon wafer, directly reaching the photodiode without passing through…
bsi image sensor, cmos image sensor, bsi process, image sensor fabrication
Backside Illumination (BSI) Image Sensors are the CMOS image sensor architecture where light enters from the back of the silicon wafer (opposite the metal wiring) — eliminating…
backside illumination sensor yield prediction, backside illumination sensor, illumination sensor yield, sensor yield prediction, backside illumination, illumination sensor
Backside Illumination Sensor Yield Prediction with Machine Learning Introduction & Motivation Backside Illumination Sensor Yield Prediction with Machine Learning addresses…
lithography
Backside lithography is the photolithography sequence performed on the wafer rear surface to pattern features after thinning or carrier bonding - it supports backside contacts…
Power Delivery, process, fabrication
Backside Metal Power Delivery Process is an advanced semiconductor manufacturing sequence that patterns metal power and ground planes on the back surface of wafers after thinning…
process
Backside metallization is the deposition and patterning of metal layers on wafer backside to create conductive, thermal, or bonding interfaces - it is a key enabler for power…
backside metal stack, wafer backside routing, backside redistribution, backside power metal, bspdn
Backside power delivery networks (BSPDN) move the power distribution grid from the front side of the wafer, where it competes with signal routing, to the back side, where it…
backside metallization stress prediction machine learning, backside metallization, stress prediction machine, backside metallization stress prediction, metallization stress
Backside Metallization Stress Prediction with Machine Learning Introduction & Motivation Backside Metallization Stress Prediction with Machine Learning addresses a central problem…
backside power bspdn, bspdn, backside power delivery, backside power, buried power rail, nano-tsv
Backside power delivery network (BSPDN) is the generic name for moving a chip's power wiring to the back side of the wafer, separate from the signal wiring on the front, and it…
backside power delivery, backside power delivery network, powervia, buried power rail, advanced technology
Backside power delivery networks (BSPDN) move the power distribution grid from the front side of the wafer, where it competes with signal routing, to the back side, where it…
bspdn, backside power, power via, buried power rails, backside power delivery network
Backside power delivery networks (BSPDN) move the power distribution grid from the front side of the wafer, where it competes with signal routing, to the back side, where it…
backside pdn, backside power rail, powervia, backside routing, bspdn
Backside power delivery networks (BSPDN) move the power distribution grid from the front side of the wafer, where it competes with signal routing, to the back side, where it…
backside pdn, bspdn, power via backside, buried power rail
Backside power delivery networks (BSPDN) move the power distribution grid from the front side of the wafer, where it competes with signal routing, to the back side, where it…