Il prossimoProssima guida
Perdita di triplette e apprendimento metrico
Fondamenti
GUIDA AI FONDAMENTALI
A loss defines the objective used to assess predictions or another training target.
Ascolta questa guidaNarrato dall'IA · ~3 min
Individual loss values are often combined into a scalar for optimization, and the chosen objective is a proxy that may differ from the application's evaluation metric.
A prediction loss compares outputs with training targets according to a specified rule. Those targets can be labels or values constructed for the learning task; they should not automatically be called unquestionable truth. Implementations may produce per-example losses and then sum or average them for an optimization step. Choosing the rule matters. Mean squared error averages squared residuals, giving large residuals more influence than mean absolute error, which averages their absolute values. For residuals one and three, mean squared error is five, while mean absolute error is two. These are illustrative calculations rather than performance results from a dataset. Classification commonly uses cross-entropy. With a one-hot target and a predicted probability p for its class, the loss is minus log p. A lower probability for the target class gives a larger loss, so confident wrong predictions can incur high loss. The logarithm base determines the units: base-two loss is in bits, while a natural logarithm gives nats. For a target probability of one-quarter, minus log-base-two p is two bits. Gradient-based optimization needs usable derivative information, but differentiability requirements should not be generalized to every learning algorithm. Some losses have nondifferentiable points handled by an appropriate subgradient or surrogate. Accuracy is a discrete evaluation metric and is generally not directly useful as a gradient-training objective. A suitable training loss supplies an optimization signal; final task quality needs evaluation using the relevant metrics.
Utilizzo della perdita di entropia incrociata per addestrare un classificatore di spam e-mail che penalizza le classificazioni errate sicure
Scegliere l'errore medio assoluto per la previsione del prezzo delle case in modo che alcune ville estreme non dominino la formazione
Applicando una perdita contrastiva in modo che un modello di riconoscimento facciale riunisca le immagini della stessa persona
Progettare un modello di ricompensa in perdita per indirizzare un chatbot verso risposte più utili e oneste
Objective design can combine prediction terms, regularization, and other task requirements. Comparisons should state the target definition, formula, reduction, units, and evaluation metric. An objective can decrease while an application metric remains unchanged or worsens, so improvement should be measured rather than inferred from training loss alone. Custom losses need examination of the behavior they encourage as well as their numerical and optimization properties.
Newsletter gratuita
Tre storie verificate sull'IA ogni mattina nei giorni feriali, scritte in un inglese semplice. Gratuito per sempre, senza pubblicità.
Un'email ogni giorno feriale. Disiscriviti con un clic. Non vendiamo né condividiamo mai il tuo indirizzo.
Mettiti alla prova
Feedback immediato su ogni risposta e un certificato condivisibile con un ID verificabile una volta superato un corso.
Sostieni l'istruzione gratuita sull'IA. AI Understanding è un'organizzazione non profit 501(c)(3) — nessun annuncio, nessun paywall, mai. Fai una donazione
A loss defines the objective used to assess predictions or another training target. Individual loss values are often combined into a scalar for optimization, and the chosen objective is a proxy that may differ from the application's evaluation metric.
A loss defines a training objective; it does not automatically measure every aspect of application quality.
L'entropia incrociata confronta le distribuzioni di probabilità previste con le etichette vere ed è la scelta standard per la classificazione.
L'errore medio assoluto tratta tutti gli errori in modo proporzionale anziché elevarli al quadrato, quindi alcuni valori anomali estremi hanno meno influenza.
This concerns gradient-based training, not a claim that every possible learning algorithm requires a differentiable loss.
The term minus log p grows as the target-class probability shrinks. This does not guarantee a particular gradient magnitude for every model weight.
Continua a imparare
Altre guide selezionate per questo argomento
Il prossimoProssima guida
Perdita di triplette e apprendimento metrico
Fondamenti